Why true innovation means not following the crowd

Attracting funding for a new drug therapy is never easy – particularly when the treatment hasn’t yet been clinically trialled. But, as one developer discovered, it can be done. Incanthera, a small UK-based oncology company, explains how...

When we set out to move a new tumour-busting chemical entity from its academic origins within the University of Bradford, we knew it would not be easy. Drug development is time consuming, costly, and given that most drugs fail somewhere in development, funding a new drug development programme is inherently risky.

But ICT2588 was designed to kill tumours without causing any nasty side effects and the in vivo data backs this up. The science is great, supported by a well thought out business plan and a credible management team – essential to overcome the risk-averse world of biotech financing. Despite the inevitable difficulties, we were confident that potential investors would be interested. In addition to the tumour-reducing element, other patient benefits include improved life expectancy, fewer side effects, and reduced frequency of treatment, resulting in potential cost savings for medical professionals and less disruption for patients. We just had to ‘de-risk’ the project as far as possible to increase its appeal to potential partners.

In science-based commercial opportunities, ‘de-risking’ is a practice regularly used by companies in order to make an offering attractive to those investors who prefer to avoid anything that is too early stage or doesn’t have a “hot” target. Potential investors who like an idea presented to them will often invite the developer to contact them again when they have more data – at least Phase I clinical safety. This approach has led to the so-called ‘Valley of Death’ for many promising drug developments. The sums required to cross the valley are relatively modest in terms of overall drug development. But new start-ups are immediately limited to a small group of funders who are prepared to invest pre-clinically.  Having a credible management team with the right networks is critical to getting the attention of this small group.  Many promising projects fail at this stage, and the ones that attract funding are those with both appropriate connections and great science.

Developers fortunate enough to secure a meeting with one of pre-clinical funders usually find that the more fashionable and ‘hot’ the drug target is, the greater the interest. In cancer drug development, for example, cytoxic agents are seen as somewhat old-fashioned even though they still represent some of the best cancer therapies available. A “hot” drug target generally equates to something that is popular, with consequent clustering behaviour around a small number of such targets. The latter appears to give both drug developers and investors comfort in that they can see others have already invested in something similar which re-enforces their belief that the development is viable. The downside to this, which seems to be largely ignored, is that you then end up with an unrealistically large number of projects clustered around the same target. This can be further exacerbated by similar clustering around accepted sets of chemistries leading to a densely populated patent space with little room for novelty or breadth of claims.

Current projects in oncology provide a fine example of the fundamentally risk-averse nature of the investment and drug development industries, with the subsequent consequences of ‘clustering’.  Recent analyses of nearly 1,000 cancer projects in the US (PhRMA 2012) show that a mere eight drug targets account for around 40% of all projects in pre-clinical or clinical stages of development¹.

Furthermore, all eight of these targets are kinases. Each of these eight targets was identified as having at least 24 individual clinical programmes, VEGF/VEGFR topping the table with a whopping 70 molecules in the clinic. Add in the known pre-clinical projects and VEGF/VEGFR alone accounts for nearly 150 projects out of the 1000 in total. This presents an issue for developers, as later entrants in the kinase race have to rely on failure of earlier entrants to have any hope of success. Being the 15^th new VEGF inhibitor to market is hardly a reason to celebrate, so the 150^th could find the task almost impossible. The consequence of needing mass failure for your own success is that your target no longer looks so hot.

Trying to sell the technology behind ICT2588 was therefore harder than it could have been because it was unfashionable as a cancer cytoxic; It was too early stage (no clinical data) and it wasn’t in the “good” target cluster. Even worse it had some ‘bad’ target associations – ‘bad’ being basically anything that has failed spectacularly or repeatedly in clinical trials.

Just saying the words “ICT2588 is an MMP targeted cytotoxic pro-drug with vascular disrupting properties” immediately induced a quasi-Pavlovian response in otherwise rational technology assessors. Many potential investors heard the words ‘cytotoxic’, ‘vascular disrupter’, ‘MMP’ and instantly seemed to implement the ‘three strikes and you’re out’ rule.

Strike 1 - Cytotoxics, as mentioned above are generally regarded as old fashioned. To a large extent so are pro-drugs, possibly through association with the much overhyped magic bullet so beloved of the popular press.

Strike 2 – Data from in vivo studies with ICT2588² shows that the cytotoxic effect manifests itself in vascular disruption within tumours. In recent years there has been a succession of vascular disrupting agents (VDAs) that have had serious problems in clinical studies due to cardiotoxic effects.

Strike 3 – Targeting MMPs (matrix metalloproteinases) has a long history of failure. Roche, Bayer Pfizer and BMS all had MMP targeting compounds that failed in Phase III clinical trials. Most significantly in this country, British Biotech progressed 5 different MMP targeting compounds as far as Phase I to III clinical trials. The fiasco that killed off British Biotech over its Phase III results with Marimastat lingers in the minds of many to this day.

So how do you convince an investor (Incanthera now has several) that a pro-drug targeting MMP and acting as a VDA is a good idea? The imperative is to focus on the outcome, stunningly good anti-tumour activity with no systemic toxicity, then deal with the objections one by one.

Scientifically, the matrix metalloproteinases are an interesting family of enzymes that are generally inactive in the healthy adult body. They are involved primarily, but not exclusively, in tissue remodelling and are naturally expressed during wound healing, for example, where they are involved in revascularisation. MMP-associated diseases include arthritis and cancer. In the latter, developing tumours express MMPs in order to remodel their local environment, permitting tumour growth. Growing tumours also require the development of new vascular tissue to supply oxygen and nutrients to the expanding tumour mass. This neo-vasculature also expresses MMPs. Previous MMP targeting drugs have all focussed on inhibition of MMP activity as a way of controlling tumour progression.

ICT2588 is different in that it does not inhibit MMPs but instead becomes the substrate for the natural activity of specific membrane bound MMPs (MT-MMPs). Enzymatic cleavage of ICT2588 by these MMP’s at the site of the cancer, releases the cytotoxic moiety. As can be seen in Fig 1, the central linker in ICT2588 is a nine amino acid peptide attached at the NH₂-terminus to FITC and COOH-terminus to azademethyl colchicine, a colchicine derivative. This peptide sequence has been designed to be specifically cleaved by MT1-MMP and is key to the localised targeting of ICT2588.

The N-terminal FITC molecule acts as an end cap, stabilising the molecule by blocking non-specific exopeptidase activity. This is crucial to the targeting of the molecule as non-specific breakdown would result in systemic distribution of the active moiety, probably leading to unwanted cardiotoxicity. The choice of FITC as an encap was almost entirely serendipitous and was originally used as FITC is designed to conjugate to peptides and is readily available in the lab. Subsequently scores of other end caps have been tried but none have proven as effective as FITC.

Colchicine, as a plant extract, has been used medicinally for thousands of years and is currently, in purified form, used to treat gout. Like many VDAs that have been developed, colchicines bind to the ?-subunit of tubulin dimers, causing disaggregation of microtubules. In dividing cells this results in mitotic arrest and subsequent cell death. In vascular epithelium, microtubules play a key role in maintaining the cellular architecture and loss of microtubular structure causes cellular collapse and loss of epithelial integrity.

Xenograft models have been used extensively to show that ICT2588 not only has excellent anti-tumour properties but pharmacokinetically is also very stable systemically with approximately 99% of all free azademethyl colchicine being found only in tumour tissues. Ex vivo work with a number of human tissues including liver and heart again confirmed the stability of ICT2588².

So far, our approach has worked. Since launching in 2010, Incanthera, which is backed by the University of Bradford where its academic founders benefit from the support of Yorkshire Cancer Research and Cancer Research UK, has secured investment from SPARK Impact, managers of the North West Fund for Biomedical. Incanthera has also raised funding from private individuals who share a concern for improving the chances of surviving cancer, and are now in the final stages of securing further investment to progress to Phase 1 clinical trials. Finally, Incanthera recently won Bionow’s Start-Up Company of the Year award.

So ICT2588 may not be fashionable or have a hot target, but maybe it shows that true innovation in drug development means not following the crowd. Certainly ICT2588 shows that MMPs are not dead as a target and that the unwanted toxicity of VDAs can be tamed. As Aristotle said “the whole is greater than the sum of the parts”.

Authors: Dr Kevin Adams, Programme Manager and Dr Simon Ward, CEO of Incanthera (simon.ward@incanthera.com)

references:

1. Booth, B (2012): Cancer Drug Targets: The March of the Lemmings

1. Atkinson, JM, Falconer, RA, Edwards, DR, Pennington, CJ, Siller, CS, Shnyder, SD, Bibby, MC, Patterson, LH,  Loadman, PM and Gill, JH (2010): Development of a Novel Tumor-Targeted Vascular Disrupting Agent Activated by Membrane-Type Matrix Metalloproteinases. Cancer Res; 70(17); 6902–12