David's New Book

Wednesday, December 21, 2016

Aminoglycosides in the News

The title of this blog almost sounds like an oxymoron.  Aminoglycosides are boring, aren’t they? And they’re old! But with the onward march of antibiotic-resistant infections, could it be that the aminoglycosides are going to become not only news but clinically relevant as well? Well . . . . .maybe.

Achaogen has announced the top line results from their pivotal trials of plazomicin, a (not so) new aminoglycoside with activity against highly resistant Gram-negative pathogens. They ran two trials.  The first was a non-inferiority trail using plazomicin compared to meropenem to treat complicated urinary tract infections. The results are shown here.
Results for FDA pre-specified composite endpoint of clinical cure and microbiological eradication in the microbiological modified intent-to-treat (mMITT) population were as follows:
   Day 5: 88.0% plazomicin vs. 91.4% meropenem (difference -3.4%, 95% CI: -10.0, 3.1%), indicating statistical non-inferiority
   Test-of-Cure: 81.7% plazomicin vs. 70.1% meropenem (difference 11.6%, 95% CI: 2.7, 20.3%), indicating statistical superiority
Results for EMA-specified endpoints of microbiological eradication at the test-of-cure visit were as follows:
   mMITT: 87.4% plazomicin vs. 72.1% meropenem (difference 15.4%, 95% CI: 7.5, 23.2%), indicating statistical superiority
   ME: 90.5% plazomicin vs. 76.6% meropenem (difference 13.9%, 95% CI: 6.3, 21.7%), indicating statistical superiority.
  These data indcaite that plazomicin is at least as effective as meropnem and may even be superior for cUTI. The trial excluded meropenem-resistant pathogens so it is hard to see why this would be so – but the data are the data.

 The other trial that Achaogen ran was the CARE trail that they discussed at the recent FDA workshop that was the subject of a previous blog. The results they reported were from patients with (apparently) known carbapenem-resistant infections treated with either plazomicin + meropenem or tigecycline vs. colistin + meropenem or tigecycline.  Thus it was a comparison of combination therapy where the aminoglycoside plazomicin was used as one part of the combination vs. using colistin. The infections treated were either nosocomial pneumonia or bacteremia. No breakdown as to the source of the bacteremia was reported and in an email response to my questions, Achaogen replied that they were still analyzing the data and had no further information to share as yet. This is a crucial question (see below).  The results are shown below. 

Plazomicin Colistin
28 day all cause mortality 4/17 (23.5%) 10/20 (50%)
or significant complications
28 day all cause mortality 2/17 (11.8%) 8/20 (40%)

Clearly the numbers here are very small. My quick statistical analysis suggests that on the first endpoint there would be a P value of around 0.07 – a clear trend towards superiority of plazomicin.  But on the second endpoint, the P value would be closer to 0.2. Yet these are the kind of data that the regulatory agencies will have to try to interpret as they get into these small and extraordinarily difficult trials. That an aminoglycoside would be superior to colistin is not surprising.  And I’m sure these clinical data will be buttressed by strong PK/PD justifications for the activity of the drug.  Finally, when seen in the light of the large trial in urinary tract infection, the case for plazomicin becomes stronger.

At the same time, Sanofi has announced that they are taking over further progression of Warp Drive’s aminoglycoside program. It seems like 2016 is the year of the aminoglycosides.  What the Warp Drive compounds look like – I have no idea.

The remaining question for clinicians is, do we really need or want new aminoglycosides? Aminoglycosides are useful in the treatment of TB and serious Enterococcal infections - but our current aminoglycosides are probably adequate there. Among Gram-negative infections, one  area where combination therapy with aminoglycosides might be important is in the treatment of Pseudomonas infections in severely immunocompromised patients. Aminoglycosides are clearly effective in cUTI caused by Gram-negative pathogens. Otherwise, the data suggesting that these compounds work as well as other classes of antibiotics in Gram-negative infections remains controversial. I reviewed some of these data in a recent blog. My own clinical experience plus data from several studies where my lab participated suggested that combination therapy of Gram-negative infections with aminoglycosides did not prevent emergence of resistance and did not reduce mortality compared to treatment with beta-lactam antibiotics alone. Yet, in these days of increasing numbers of infections with highly resistant pathogens where colistin is the only available therapy, an active aminoglycoside might be a welcome alternative. Opposing this inclination is the market entry of other beta-lactam antibiotics active against highly resistant pathogens such as Avicaz and others in the late stage pipeline might that make these new aminoglycosides somewhat less relevant. Analysts have estimated a $340 million peak year sales for plazomicin.

Saturday, December 10, 2016

The Threat to FDA

As many of you know, I have had my “disagreements” with the FDA over the years.  But I have never questioned the fact that without the FDA, we would all be in a world of trouble. Just look at what’s happening with supplements and you’ll begin to understand what I’m talking about.

The very beginnings of the FDA were in the patent office in 1848. So-called “patented medicines” were being sold throughout the US as cures for whatever. A number of high-profile poisonings from tainted products had occurred including several affecting soldiers in the army. The patent office was given the task of identifying potential poisons among products being sold to the armed forces. In 1862, this effort was transferred to the newly created department of agriculture.  There, responsibility for the safety of food was also added. Chemical additives to preserve food included a number of poisons as well. These inspections fell to the department of chemistry within the department of agriculture.  By the end of the 19th century, over half of all newspaper advertisements were for nostrums claiming to cure everything from cancer to rheumatism. One chemist, M.J. Bailey, reported,"More than one half of the most important chemical and medicinal preparations … come to us so much adulterated, or otherwise deteriorated, as to render them not only worthless as a medicine, but often dangerous."  Europe started embargoes against American drugs and food. 

In 1905 Sinclair’s The Jungle was published.  The conditions described in slaughterhouses were so disgusting that Teddy Roosevelt decided to send inspectors to Chicago.  They found what Sinclair had described. The first food and drugs act was passed in 1906.

After passage of the law, the maker of Cuforhedake Brane-fude – claiming that this product provded certain and harmless relief and contained no poisons whatsoever – was charged under the new law for selling a product with acetanilide – a known poison that was responsible for at least 22 deaths. The product had brought in $2 million. The drugmaker was fined $700, changed the label and went on making the product for years.

The most famous case occurred in the 1930s when one of the new sulfa antibiotics was sold tainted with polyethylene glycol. Over 100 people died, many of them children.  But the company had broken no law. The label was not “misleading.” This, more than anything else, led to the passage of the Food Drug and Cosmetics Act of 1938. Proof of drug safety would be required before a drug could be marketed. But the 1938 law said that companies had to submit data showing safety and the FDA had 90 days to object before the drug would be marketed.

Then came thalidomide and thalidomide babies.  That led to passage of the Kefauver-Harris Amendment of 1962 – signed by JFK. This law required the demonstration of both safety and efficacy setting basic standards for what evidence could be considered. While the FDA has progressed since 1962, this law remains the basis of everything they do.

Today, some want to turn the clock back to 1938.  In particular, Jim O’Neill (no relation to the Jim O’Neill who led the UK’s Antimicrobial Resistance task force) wants to have drugs approved once they have shown to be “safe” and then have some sort of rolling approval where there would be an obligation to show efficacy. Not only would this be a terrible step back into a sad and destructive history, but it is completely irrational.  The concept of safety can only be understood in the context of benefit.  If the benefit outweighs the risk, a product can be considered safe even if there are some potential liabilities.  This has been our entire approach to the treatment of cancer and other serious diseases for decades. O’Neill is apparently under consideration for Commissioner of the FDA.  Are you kidding?  He is not a physician nor is he a scientist.  He is a Managing Director at Peter Thiel’s Mithril Capital Management.

Although to achieve what O’Neill would like, laws would have to be changed and probably the majority of people at the FDA would simply walk out if this were to occur. So, it seems unlikely that he would succeed in this particular quest.  But someone like him could do a great deal of damage.  Why would we ever want to go through the early 1900s of drug regulation again? Please – let’s not go there.

Tuesday, November 29, 2016

The 21st Century Cures Act and Antibiotics

It looks like congress is finally going to pass the 21st Century Cures Act. Its 996 pages long!  Please read it yourself (just kidding). Of course, this doesn’t even come close to France’s labor laws at 3500 pages – but its still impressive. The bill contains over $6 billion in funding most of which will go to NIH for work on opioid addiction, cancer and Alzheimer’s disease. There is no doubt that the NIH needs more money since funding has been flat for years – therefore, corrected for inflation – we’re on a downward spiral. But targeted funding like this tends to take money and attention away from other areas of need – like antibiotic research for example.

So – what’s in it that’s good for antibiotics? Lots!

Limited Population Antibiotic Development (LPAD). There is a clear pathway spelled out for the FDA on antibiotics for use in limited populations where approval is based on smaller data sets than would normally be expected.  Of course, we all know that the FDA is already doing this under existing regulations – but this puts Congress on record as supporting this approach.  The FDA apparently thinks that this law will help them do more to bring new antibiotics to market quicker – and if the FDA thinks this will help then so do I. (OK – I changed my tune . . .). The best example of what the FDA can do is their approval of cetazidime-avibactam based mainly only on phase II data.  Hopefully, more drugs can now be approved more quickly in this way.  One also hopes that pathways will now appear for the rapid approval of pathogen-specific antibiotics (see my last blog on this).

Novel Clinical Trial Designs.  The Secretary HHS is ordered to convene a public meeting providing input to the kinds of novel clinical trial designs and statistical methods that would be required to implement LPAD.  This would then be followed by guidance from FDA in this regard. I suspect this is already in the works and the infectious disease group has already held public meetings on this.
The role of BARDA is clarified and expanded. But BARDA probably still does not have the funding to provide a business model for pathogen-specific antibiotics.

Antimicrobial susceptibility testing.  The FDA is ordered to establish a website with interpretive criteria for susceptibility testing that will be frequently updated and that will start upon approval for each newly approved drug.  The FDA will have the authority to rely on outside organizations for help here – similar to the way this works in Europe. I don’t see anything here that will speed the availability of automated testing to laboratories and physicians in the US. This is still desperately needed and I don’t see a way forward here as yet. The most important aspect of this part of the Act appears to be the updating of information on already marketed antibiotics. But that, in and of itself, is a major advance.

Breakthrough Diagnostics.  There is a section on the rapid development and approval of new diagnostic devices.  This is an important section that might make it more feasible to bring new rapid diagnostic testing devices to the US market.  We have been way behind Europe in this regard.

Recruitment and Retention at FDA.  There is an entire section spelling out the skills that will be required for new hires in the clinical review roles at FDA. This includes provisions for justifying higher salaries and other areas that would allow more efficient recruitment and better retention of FDA employees.  This is obviously needed based on what I have been reading. I didn’t see anything there specifically on conflicts of interest and making exceptions although the Secretary of HHS is to report back to congress as to needs in this area.  There is also nothing related to getting more expertise on advisory panels by waiving some conflict of interest criteria.

As I noted, there are obvious areas where the bill could still be improved. Hopefully the bill will pass without amendments to defund Planned Parenthood for example.  

My congratulations and thanks to Pew Charitable Trust and to the Infectious Diseases Society of America for their tireless efforts to get this passed over the last 3-4 years.

But, at the end of the day, until we have a business model that works for antibiotics, our pipeline will remain dry in spite of this legislation.

Sunday, November 20, 2016

Pathogen-Specific Antibiotics - Business Models

I just returned from giving a talk to microbiologists on the issues surrounding the clinical development of antibiotics with activity against specific pathogens – that is – the opposite of so-called broad-spectrum drugs. I used as an example an antibiotic that targets Pseudomonas aeruginosa, Polyphor’s POL 7080.  

Microbiologists play an important role in constraining the use of antibiotics in both the hospital and community settings.  They are interested in the development of antibiotics that are more specific and therefore less likely to engender antibiotic resistance as collateral damage.  What I mean is – when treating someone for a Pseudomonas infection with a carbapenem antibiotic, you may select for resistance in Klebsiella since the carbapenem kills indiscriminately.  At least with a drug that targets only Pseudomonas, you only have to worry about selecting resistance in Pseudomonas.

I pointed out that, at least today, from a regulatory point of view, we still don’t know how to run a feasible trial that will result in marketing approval for such a drug. While I am confident that this problem will ultimately get solved, I am not so sure about the business model for such a drug.

The problem for this sort of antibiotic is that it will be used, at best, only sparingly. From the stewardship point of view – this is the ideal situation. But, if the company marketing the product has to depend on sales volume, it will be even more difficult to provide a return on investment for this sort of antibiotic compared to a broad-spectrum agent that can more easily be used on an empiric basis. Since companies are not in the business of providing for the public health in some sort of charitable way, why would they spend their limited resources on developing this sort of product? We, as a society, need to provide a business model that will work.

In Europe, where most countries already have a form of socialized medicine, the models being discussed all include some sort of market entry reward. In all these models, when a company gets approval to market an antibiotic that meets pre-determined criteria, they receive a payment of some sort.  How this payment is distributed is the basis of the debate on which model to use. Regardless of the model, the payment would be such that it would provide a return on investment for the company. While such a model was about to be put into use, I understand that everything is now on hold awaiting a better understanding of what Brexit will mean for payers in Europe. 

Here in the US, the model that seems to be preferred by people discussing this issue is the so-called patent voucher. In this system, a company that successfully brings a needed antibiotic to market would be rewarded with an additional period of exclusivity on the product of its choice from its portfolio of marketed products.  If, for example, the company were Gilead, it might choose extra time of exclusivity for one of its multi-billion dollar Hepatitis C drugs. This would guarantee a return on investment for the new antibiotic even if it were only used for a small number of patients every year.  The last time this model was discussed by the lobbying group from the Infectious Diseases Society back in 2004-5, the model was dead on arrival in congress partly because of the blowback from both consumers and the generics industry.  How this model will fare under a Trump administration is impossible to predict – but past experience on the Hill is not encouraging.

Some companies, like Merck, have already staked out a position here. Merck seems to want to be able to charge an appropriately high price and have that price be covered by various payers globally.  Merck and other large pharmaceutical companies might or might not be encouraged by the experience of Allergan and ceftazidime-avibactam in the US. Ceftazidime-avibactam or Avycaz was approved in the US in 2015 based on phase II data.  This antibiotic is the only alternative to the toxic and poorly efficacious colistin/polymyxin for the treatment of certain highly resistant Gram-negative infections. These infections remain, happily, relatively uncommon in the US. The label it received from the FDA restricted it to use only when other alternatives were not thought to be available.  The price was set at $12,000 per course of therapy – but payers actually provided something like $8500. This makes it the most expensive antibiotic ever marketed globally. Analysts predicted $300 million peak annual sales for the drug.  The first half of 2016 Aycaz had only $22 million in sales with the second quarter coming in at $13 million. Based on second quarter data and accounting for growth, this year could have seen $60 million.  Two events have altered this landscape.  First, the label for Avycaz has changed and was expanded now that they have an approved sNDA based on their phase III data. How this will affect the price is not yet known.  Secondly, Allergan has experienced manufacturing problems from the GSK supplier that will slow down sales in the latter half of this year.

My own opinion is that we need something beyond price and that Allergan’s experience makes a strong argument for that view. Allergan will now struggle for several years to achieve the analysts’ peak annual sales forecast of $300 million if it ever does so. And this level of sales, albeit it is only in North America, may not be enough for many companies in any case. Other pharmaceutical companies will be watching this closely and they may not be encouraged.  But we need more large companies to join the fight against antimicrobial resistance through the research and development of new antibiotics active against resistant pathogens. The experience of Allergan coupled with continued dithering on the Hill does not bode well for our goal of bringing more large companies into the field.