At the outset of this piece, I am going to make a clear disclaimer, namely that I in no way imply that Novogen’s (ASX : NRT) Cantrixil product is either excessively toxic or unsafe – or conversely that it actually works. The purpose of this piece is to explain – in a high level way – why Novogen’s story around Cantrixil doesn’t add up in terms of progressing the program, with a specific focus on toxicology. I have elected to write this post at the request of several readers who have asked me to elaborate on my negative comments regarding Novogen’s progress with a pre-IND tox package.
Toxicology studies are a fundamental part of drug development and depending on the type of drug being developed, toxicology can be more or less complicated. The good news is that for cancer drugs – and particularly small molecules like Cantrixil – toxicology really isn’t all that complex and the regulatory requirements are very well understood. Cancer drugs have an advantage in that they are not usually tested on healthy volunteers, and because patients are seriously ill and often post-reproductive years, a lot of the long-term risk factors of a drug (i.e. carcinogenicity, genotoxicity, impact on reproductive health) aren’t terribly relevant and don’t need to be tested right away. The general philosophy is that there is no better “model” for testing a cancer drug, than a cancer patient, and the regulators don’t actually make it all that difficult to do so.
This is particularly the case in Australia where it is quite straightforward to try a proof-of-concept for a new cancer drug in patients via the TGA’s CTN scheme. In general, in most parts of the world, relatively minimal toxicology studies can pave the way for small “first in human” (FIH) clinical trials that may elucidate whether a drug has any potential anti-tumour effect, and whether the drug might be tolerated by a patient. In Europe, where there has long been a fairly general “acceptance” that animal toxicology studies have a limited predictive value of effect in patients, it is possible to dose cancer patients on the back of a very basic tox package – like a single-species studies in rats (the US is slowly moving this way as well – with growing recognition that dog studies add little value). This is fairly fast and cheap to do – a few months, and a couple of hundred $k, depending on the dosing schedule anticipated in patients.
This leads to the basic question : What is a toxicology study in the context of a cancer drug?
Well, basically it is a study in animals to determine something called “Dose Limiting Toxicity” or DLT. Although DLT sounds conceptually simple, it can actually be very complex and a whole range of measurements can be used to determine DLT, depending on how the drug works. For example, if a drug has a lot of kidney toxicity, then the measurement of kidney function biomarkers might establish limiting toxicity. If a drug depletes a patient’s immune cells or kills bone marrow, than circulating immune cell count in a blood sample might be used to determine a safe cut-off point. In pre-clinical (non-clinical) studies before going into patients, toxicity studies combined with mechanism of action studies in “models” will hopefully establish the dose range of where the drug is both efficacious (i.e. makes a tumour go away, or grow less quickly) and “tolerable” by the animal. When DLT “goes” into people, it is generally described as a “Maximum Tolerated Dose” (MTD). This is because the tolerability characteristics have a big impact on whether a drug will be successful or not. Cancer drugs fail not just because of a lack of anti-tumour efficacy, but because side-effects might be really awful – even so awful that the patient can’t finish a course of therapy.
And so what of Novogen and Cantrixil?
Well, obviously companies have to beat to their own drum, not that of guys like me. Novogen obviously cares deeply about the opinion of their shareholders, otherwise they wouldn’t periodically engage in the kinds of quasi-science disclosure stunts that they do. At least the last time they did it, it wasn’t labelled as price sensitive. Because it isn’t. The most fundamental reason I remain contemptuous of Novogen is because before you declare something a “wonder drug” you should at least know whether it fulfills three basic in vivo (i.e. in a living subject) criteria. These characteristics apply to pretty much all drugs but the nice thing about developing a drug like Cantrixil is that the list of “must haves” in a clinical candidate can be very short if you plan to evaluate a new drug in patients with aggressive late-stage cancers – like ovarian cancer.
Criteria 1 : does it have a basic anti-tumour effect that supports the mechanism of action in a relevant model(s)?
Criteria 2 : does it stick around long in body through a meaningful / relevant dosing schedule to deliver that effect? By this I mean do you have to pop a pill every 20 minutes for the drug to work or can it be a once a day / once a week treatment? (noting that Cantrixil will be directly administered into the body cavity).
Criteria 3 : At the dose / schedule that the drug has an anti-tumour effect, is it reasonably safe? I say “reasonably” because we are talking about super-sick patients and the risk-benefit ratio is very different.
I believe that Novogen’s (and collaborators) pre-clinical studies suggest that 1 and 2 have been reasonably achieved. I accept this at face value, despite a dearth of detailed peer-reviewed publications and the fact that most ovarian cancer animal models are only a very rough approximation of reality. But safety – measured through toxicology and initial clinical experience – it not something that you “observe” it is something that you measure and quantify relative to dosing limits. If you are a proper pharmaceutical company, you don’t usually brand a drug until it is in Phase III – for a whole variety of reasons and I have talked about this in the past. But you certainly don’t tout the clinical and commercial potential of a drug until it passes the most basic muster, like knowing whether it has a shot of being safe in patients.
All clinical development is tough, but evaluating a new cancer drug in patients with advanced disease is a very achievable goal and it’s not terribly expensive. For a proof-of-concept study for Cantrixil, either stand-alone or in combination with another cytotoxic agent, a reasonable toxicology study (even a 2 species – rats/dogs – study to satisfy the US FDA) can be completed in 6 months and a few hundred thousand dollars. Novogen really has no business making claims about drug toxicity/safety until it has done this work and disclosed the acceptance of those studies by a regulatory agency like the FDA in the form of a “notice of allowance” for an investigational new drug (IND) application.
However, the really offensive thing for me about Cantrixil is that the company just can’t get its story right. Novogen’s consistent positioning of Cantrixil as the “first peritoneal cancer agent” in ovarian cancer is just blatantly incorrect. Peritoneal delivery of chemo is the standard of care in ovarian cancer patients.The company has also claimed that it is the first drug to be effective against cancer stem cells, also rubbish. But worst of all, Novogen makes claims that the drug kills cancer cells and in the same breath explains that it has no toxicity. If it has no toxicity, then there is probably no need for intra-peritoneal delivery, and the claim of “no toxicity” can only be made when the evidence is in place to make such a controversial statement (we have been waiting for over a year). Why is this controversial? Imagine that you have a patient with late-stage cancer, who is riddled with tumours and malignant ascites – there are tumours in the major organs, in the lymph nodes. In the case of ovarian cancer patients, these tumours can be several cm in size at Stage III+ disease. Those tumours have blood supplies, they infiltrate connective tissue, they punch holes in organs… and so do you think that a drug that selectively kills cancer cells means that it has no adverse effect on the patient?
The evidence that Cantrixil is anything more than just a story, in my opinion, remains elusive. Until the toxicology work is done, is accepted by the FDA (or equivalent) and the first patient experience reported, Novogen has no business expounding the seemingly miraculous virtues of its drug.
Note on feature image. I found the basic image doing an image search on mouse toxicology (it was on a Monsanto “hate” page) but I have not been able to find it again. Apologies for not being able to attribute it appropriately as it is rather a good drawing…