Cancer Complexity
Clinical development is outmoded; we are witnessing serious attrition in cancer R&D. Perhaps it’s time to live and die by the mantra: “no biomarker, no trial”.
What is the biggest challenge that we face in our mission to bring new, effective medicines to patients? In my view, it’s attrition. In 2004, Kola and Landis reported that only one in 20 new chemical entities (NCEs) being developed for cancer make it to the market (1), with other indications faring somewhat better. A more recent analysis from 2014 suggests that this picture hasn’t really changed and, if anything, it has become worse (2). The success rate is frankly abysmal. So, what can we do to reduce attrition rates? What are the challenges and how do we overcome them? There is no quick, easy fix. We need a major transformation in the way we develop drugs, spanning preclinical development, how clinical trials are designed, and the way in which drugs are currently approved. It isn’t possible to cover every aspect here, but I’ll give you a few ideas to get you thinking.
Developing new treatments for cancer can be potentially more challenging than for other indications for numerous reasons. There are more than 200 different cancer types, characterized by inherent heterogeneity both between and within patients. Cancers also rapidly develop drug resistance and can evolve to evade natural immune surveillance. Given such complexity, the high rate of attrition perhaps isn’t surprising. Lack of clinical efficacy is the key factor leading to attrition.
To address this, I believe that we need to push vital decision-making points back along the clinical development pathway and base them on a thorough understanding of tumor biology and pharmacology of the experimental agent. Incorporation of robust pharmacodynamic biomarkers to demonstrate target modulation in cancer and/or molecular sub-group specific expansion cohorts need to be applied in the Phase I setting. Only then can we be confident that the agent being tested is doing “what it says on the tin”. This strategy should increase the probability of downstream success and rule out agents that have no signs of biological activity or clinical benefit. In other words, we’d be living by the mantra, “no biomarker, no trial”.
The development of predictive biomarkers to enable patient stratification can help tackle the inherent heterogeneity of cancer. Initiatives such as the Genomics of Drug Sensitivity in Cancer (www.cancerrxgene.org) are helping to make this a real possibility. Adaptive early phase clinical trial designs that incorporate pre-planned analysis based on biomarker defined sub-groups, such as that of the TOPARP-A study, could also be beneficial (3). Rather worryingly (but perhaps not surprising), an analysis of discontinued oncology drugs from 2013 revealed that none of the agents terminated during pivotal trials incorporated molecular stratification markers (4). The “one size fits all” approach, which served us well during the chemotherapy era, is an outmoded clinical development strategy.
Effective drug combinations can tackle resistance by increasing the efficacy of cancer drugs, but individual companies usually only have the resources to explore a fraction of the possible drug combinations that have a valid scientific rationale. If we work together, through cross-company and academic–commercial collaborations, then we will be able to accomplish so much more. Indeed, spurred on by the promise of the new generation of immune checkpoint inhibitors, pharma companies are teaming up with other organizations to further optimize these agents with other therapies from either the same or different classes. Cancer Research UK (CRUK) together with the UK Experimental Cancer Medicine Centres are also leading the combination charge, with an initiative that aims to accelerate and broaden the number of rationale combinations being tested in the clinic by bringing together academic experts and cross-company collaborations (www.ecmcnetwork.org.uk/ca).
To tackle attrition head on, we need to raise the bar. Critical decisions need to be made earlier in the drug development process, which is a key aim at the CRUK Centre for Drug Development, and we need drug development “ecosystems” involving industry, academia and the regulatory authorities. One glimmer of hope is the fact that 2014 was a record year for drug approvals by the FDA – only time will tell, but perhaps some of the necessary seeds of change to overcome attrition have already been sown.
- I. Kola and J. Landis, "Can the pharmaceutical industry reduce attrition rates?" Nat. Rev. Drug Discov. 3(8), 711-715 (2004).
- M. Hay et al., "Clinical development success rates for investigational drugs," Nat. Biotechnol. 32(1), 40-51 (2014).
- J. Mateo et al., "DNA repair defects and antitumor activity with PARP inhibition: TOPARP, a phase II trial of olaparib in metastatic castration resistant prostate cancer," American Association for Cancer Research Meeting, Pennsylvania, USA (April 2015).
- R. Williams, "Discontinued in 2013: oncology drugs," Expert Opin. Investig. Drugs 24(1), 95-110 (2015).
James says that he wants to live in a world where the word “cancer” no longer holds any fear, and he has dedicated his professional career to achieving this aim, having been involved in cancer drug discovery and development since 2001. His experience has spanned the entire continuum from early discovery through to pivotal clinical development and he is currently the drug development scientist at Cancer Research UK’s Centre for Drug Development. The remit of the CDD is the translation and early clinical development of new anti-cancer agents covering everything from small molecules to immune and cellular based therapies.
