Manhunting high-value targets in cancer combat
Posted 15th November 2017 by Jane Williams
Cancer poses a formidable challenge to public health in modern society. In duelling rituals, the challenged party is usually given the choice of weapons, and possibilities are numerous. Knowing the weaknesses of your opponent is key to combat strategies and weapon selection.
Cancer, as defined by Monahan & Weinberg’s characterising hallmarks, is a collective term for countless variants of the disease. This challenging health threat can be pictured as an undermining, flexibly organized criminal organisation or a guerilla war enemy, the type of opponent that you wouldn’t defeat with a single precision shot, nor by dropping a devastating bomb. The complexity of the challenge was unequivocally recognised by public opinion when U.S. President Nixon signed the National Cancer Act of 1971 and the media reported his declaration of the ‘war on cancer’.
Achilles heels?
Nonetheless, metaphorical expressions such as “targeting the Achilles heel of cancer” recur in the press, suggesting the match decisive unmasking of a unique weak spot in our enemy. Among the alleged Achilles heels we find a puzzling diversity of vulnerabilities. One might recall DNA polymerases POLB and POLG in MSH2 and MLH1 deficient tumours respectively.
In another instance, a turning point for future therapies was envisioned in the analysis of a tumour’s proteasome load vs. proteasome capacity, new parameters that would predict the responsiveness to treatments with proteasome inhibitors. Moreover, tumour antigens, cancer cell flags that would train the immune system to attack them, were as well seen as soft spots, opening the door to immunotherapy on some of the trickiest cancer types.
More recently, the treatment of resistant mesenchymal cancer cells with GPX4 peroxidase inhibitors was reported as ‘breaking the cells’ armour’, blocking the resistance mechanism that prevents cell death by ferroptosis. In an ever growing list, the MCL-1 pro-survival protein appears as a targetable cancer weak spot, since a promising inhibitor (S63845) reportedly proved highly effective in treating difficult breast cancers in combination with chemotherapy.
Drivers, passengers, ringleaders, deputies, assistants, bystanders…
Genome sequencing revealed that a multitude of gene defects may co-exist in cancers of various types. While most mutated genes have no relevant functional role in the full-fledged disease, others are drivers and contribute to the cancer’s evolutionary fitness and clonal development.
While some driver mutations such as RAS are ‘broad spectrum’ with an impact on many cell types, others have a restricted influence or are selectively expressed. Cancers develop dynamically, adapting to selective pressure, with different players (genes) assuming and losing roles over a time course, be it leading roles, supporting roles or backups in organized, robust networks. Neutralising key players having their origin in driver genes resolves critical situations in certain types of cancer.
On the other hand, the selective pressure of a specific drug acting on a driver provides growth opportunities for clones that are functionally independent of the driver. Cancer cells re-model by virtue of their microenvironment. This becomes more evident as tumours spread, delocalise and intertwine with healthy tissue.
Unconventional warfare
Given the complexity and the dynamics of the oncogenic mechanisms, all options in the choice of weapons should be considered and weighed in comparison. Unraveling the convoluted mechanisms of cancer biology is a long-term objective whose pursuit is rewarding all along the way, as new targets and their role are elucidated and their clinical relevance is appreciated.
Biomarkers defining sensitive patient populations, and molecular measures of therapeutic mechanisms are tangible values in clinical practice. But the patients’ need for new drugs with better efficacy and tolerability cannot wait for a complete insight as to the means and ways cancer works. Innovative, unconventional options are to be added.
Interfering with the drivers is only partially satisfactory and cannot suffice for the future. It has been the preferred way forward in anticancer drug discovery for the last decade, as it falls within scholarly pharmaceutical research schemes. It allows projects to work on a single target starting with simple models, following a linear path of compound performance optimisation guided by structure-activity feedbacks. Unfortunately, in the overall picture of cancer cures, the focus on driver dominated diseases is quite limiting, considering also the risk of resistance development. As models closer to actual clinical pathologies are within reach, better predictions on patient responses are expected.
Taking on the challenge
The time is right to take on the challenge with unconventional means, to step out of the comfort zone, forego the pursuit of all-round solutions and bring to fruition new chemical entities with imperfect profiles, perhaps flawed with unplanned secondary activities and development issues, but representing useful and timely additions to the oncologist’s weaponry.
The running battle against cancer asks for melee weapons that can be used under defined circumstances, in alternation or combination regimens and preferably on selected patient populations. This change of pace was pioneered years ago and has taken shape in discovery approaches breaking the boundaries of medicinal chemistry rules often applied too restrictively.
Chemical space beyond the ‘rule-of-five” and the definition of drug-likeness has been explored successfully. Reactive enzyme inhibitors, covalently binding their target, have been developed. Also, rule-breaking synthetic conjugates recruiting cellular components to trigger target degradation have shown remarkable cell permeation and in vivo activity, thus exemplifying a new principle of chemical knockout of gene products. A new generation of cancer-focused biotech companies embraced synthetic lethality to leverage on a combination of targets simultaneously, thus accepting the burden and risk of more complex clinical trials in exchange for efficacy on new territory.
DNA repair inhibitors and immunotherapy, but also cancer metabolism targets are under investigation as combinable modules with improved safety over cytotoxic chemotherapy combinations. Target validation (and invalidation) is becoming a much more vast data interpretation exercise since target interplay and metabolic contexts are included in the reasonings. The comfortable monogenic disease models are being replaced by cellular systems closely mirroring the in vivo state of the pathology.
In terms of drug discovery, target-centric research needs stronger complementation with compound-centric efforts, capitalising on polypharmacology that until recently was diligently avoided. The acknowledgement that hitting a controlled number of targets simultaneously may offer solutions to pressing medical needs entails readiness to embark either in combination studies of selective compounds or in fostering the ability to optimise the performance of individual compounds on multiple targets. The latter requires deviation from the preferred mainstream structure-guided design based on predicted affinities to a single target. Both approaches bear complications in dissecting and monitoring the mechanism of action of the biological effect.
Multiparametric analyses of bioactivity and multifactorial optimisation of new chemical entities may soon become routine work in the hit-to-lead stage, especially in oncology research, which seems most penalised when practised on simplified models that invariably misrepresent the real challenge posed by the disease.
This all opens exciting opportunities for modern medicinal chemists to widen their expertise and venture into using artificial intelligence and evolutionary algorithms in areas of relevance to molecular design and chemical synthesis, building on the methodologies developed in the nineties to guide through successive generations with improved characteristics until an optimal, or near-optimal solution is obtained.
Eduard Felder is Director, Head of Chemical Core Technologies at Nerviano Medical Sciences. He will be speaking at the Global Medicinal Chemistry & GPCR Summit this month.
View the agenda to see who else is speaking.
Leave a Reply