High prices for antibiotics - a good thing

The problem of antibiotic resistance is getting another profile boost with its inclusion in the NYT's "Room for Debate" series. There is actually little debate in the section, save the attempt by the Pork Council to downplay the threat of overuse of antibiotics in the livestock industry. Instead, there are good, but somewhat unconnected pleas for test-driven use of antibiotics (David Gilbert), better financial incentives for antibiotic R&D (Brad Spellberg), a more rigorous, top-down approach to infection control (John Bartlett), and advocacy of phage therapy (Matti Jalasvuori).

I'd like to connect the dots by suggesting that most of our problems with antibiotics have an underlying root cause: that antibiotics are too cheap.

Cheap antibiotics do not produce a return on investment for R&D. Cheap antibiotics are too readily prescribed for minor respiratory infections that are predominantly viral. Cheap antibiotics can be used as a growth promoter for livestock, and as a substitute for clean living conditions.

If a life-saving course of antibiotics cost thousands of dollars, rather than hundreds, pharma companies would race to develop them. If Z-packs cost hundreds of dollars, doctors would not be so quick to prescribe them for sore throats. If it cost tens of dollars, rather than pennies, to dose hogs with tetracyclines, then routine use of them would cease.

Antibiotic susceptibility in bacteria is not precisely a finite resource - it's not as though there are only a certain number of prescriptions that can be written before a given drug becomes useless. But it is certainly a common resource, one that can be depleted rapidly or husbanded for the use of future generations. We have been choosing to deplete their effectiveness rapidly, but we don't have to.

We could preserve effectiveness by a top-down approach, placing limits on who could prescribe antibiotics, or requiring test results for prescription. But these regulations would quickly become outdated as new technologies and diseases emerge. Regulation also would provide no financial incentive for R&D investment.

In contrast, high prices would make development of narrowly-targeted antibiotics more attractive, and would rein in overuse in both humans and livestock. Although individuals might pay more, society and the health care system would pay less, as there would be fewer patient deaths and fewer long stays in the ICU due to untreatable infections.

How would we obtain high-enough prices for antibiotics? Cancer drugs often cost many tens of thousands of dollars for a course of therapy. It's not clear how they obtained this pricing power, other than the fear of cancer that our society has cultivated. We've lost our fear of bacteria, but starting a scare campaign to restore it hardly seems ethical or feasible.

A conservation tax would raise prices to end-users, but would not provide a ROI for developers, unless the tax proceeds were channelled back to them. This system would be susceptible to abuse. Another approach might be to set a floor on prices, much as we do for milk and other commodities. A guaranteed price would reduce market risk for developers, as well as suppress overuse of currently available antibiotics. A price that is determined by trends in the development of resistance would make this system more effective, and somewhat less arbitrary and prone to abuse.

Higher antibiotic prices would reflect their true value to society. They would let us get to a future where new antibiotics are being developed, while old ones maintain their usefulness. It's a step we need to take.

 

Posted with Blogsy

Empiric therapy - not a substitute for rapid testing

In the last post I wrote about the case for the value proposition of narrowly-targeted antibiotics as developed by Spellberg and Rex. Their model has an implicit assumption - that the hypothetical high-priced antibiotic will be prescribed in a timely manner to patients who will benefit from it, and only to those patients. My point in that post was that a companion rapid diagnostic would be needed in order to realize this scenario, that such a test is likely to have a significant cost as well, and that this cost needs to be factored into the analysis.
David Shlaes has taken issue with this view, explaining that

"Luckily, physicians don't wait for the diagnosis before treating. They treat empirically - and if they are in a hospital where MDR Acinetobacter is a risk - they will use a new expensive drug to cover for that possibility until they have confirmation or not. "

He is of course correct. No physician faced with a failing patient would withhold a drug that might have a significant therapeutic benefit. But this practice has all sorts of implications for the value of narrowly-targeted antibiotics, and none of them are good. I'm not sure that "luckily" is the adverb I would have chosen to describe the situation.
Let's be clear - empiric prescribing is a necessary response to imperfect information. But there should be no illusions that physicians have some magic powers that make empiric prescriptions much better than guessing. Indeed, a careful study of prescribing practices is entitled "Empirical antimicrobial therapy for bloodstream infections ...: No better than a coin toss". As the title suggests, when faced with a binary choice between choosing an antibiotic that is optimal for MRSA and one that is optimal for MSSA, doctors did no better than predicted by chance. There is no reason to think they would do any better in choosing optimal antibiotics for Gram-negative infections.
Combine this element of chance with the fact that the organism of interest in this scenario is not all that common, and the case for the value of an antibiotic narrowly targeted toward it can deteriorate pretty quickly. Acinetobacter infections are not currently likely to constitute more than 10% of Gram-negative infections in any hospital in North America. Although not dominant, this level is high enough to be a threat in any patient with a serious infection, and physicians will have to treat accordingly.
If they prescribe the antibiotic to cover possible Acinetobacter infections in half of these patients when only 10% are actually infected, then there are several unintended consequences, all of them bad (except for the 10% who actually have Acinetobacter infections).
First, the value proposition for the antibiotic becomes far weaker. In Spellberg and Rex's median case the antibiotic costs $10K per dose, and the net cost for an additional year of life saved is only $3K (anything less than $50K is considered to have a favorable cost-benefit ratio). But if the antibiotic has no benefit for the patients who turn out not to have an Acinetobacter infection (and this is stipulated in their scenario), then use of the drug is a dead loss economically. If 50% of patients receive the antibiotic when only 10% are infected with Acinetobacter, then the cost per year of life saved goes from $3K to $15K, a much less attractive proposition.
Second, the non-Acinetobacter patients are exposed to the risk of adverse events, with no corresponding clinical benefit to compensate. Since this is a hypothetical drug, there is no point in trying to quantify this risk, but it is surely not zero.
Third, this level of overuse will accelerate the development of resistance to the drug. Although narrowly targeted by definition, it is biologically implausible that it would have no negative effect on the fitness of any of the thousands of species of bacteria that comprise the human microbiome. And once resistance develops in one species, it is only a matter of time until it spreads to many species, including the targeted pathogen.
Now imagine that there are several narrowly targeted antibiotics available: one for Pseudomonas, another for carbapenem-resistant Enterobacteriaceae, etc. In the absence of a rapid test, how will a physician decide which to prescribe? Should she prescribe all of them? If she does, all of the negative effects of inappropriate therapy listed above would be compounded.
All of these problems go away if there is a rapid test. If I were a pharma exec, I would not begin considering a program to develop a narrowly-targeted antibiotic unless I felt pretty confident that a companion rapid test could also be developed. The path to clinical and economic value is just too full of land mines otherwise.

Posted with Blogsy

Can't have the drug without the test

As I've written before, antibiotics are too cheap. Drug companies simply can't get sufficient ROI on a product that is used once and costs a few hundred dollars. They would much rather develop drugs that patients take every day for the rest of their lives, or develop cancer medications that bring in $50-100K for a course of therapy.

Via David Shlaes, I see that Brad Spellberg and John Rex have looked into the question of how much a course of antibiotics could cost, and still return a net benefit to all parties. Their hypothetical drug is narrowly targeted toward carbapenem-resistant Acinetobacter baumanii (CRAB), a bug for which all current therapies are ineffective or highly toxic.

They find that at $10K for a course of therapy, an Acinetobacter-specific drug which reduced mortality from 20% to 10% would end up costing some $3000 per year of life saved. The comparable value for Avastin, a best-selling cancer drug, is $168,000. In other words, even a $10K antibiotic is actually pretty cheap.

What Spellberg and Rex don't address is how that cheap/expensive antibiotic gets prescribed in the first place. Since the hypothetical drug is narrowly targeted, there first needs to be an identification that the infectious agent is indeed Acinetobacter, and that it is in fact carbapenem resistant. Standard hospital micro lab procedures can do this, but it typically takes 3 days to get a result. In that time the patient is being treated with ineffective antibiotics, and may well be past the point of recovery by the time the correct diagnosis is made.

The obvious answer - and I know Brad is an advocate of this - is to develop rapid diagnostics that would identify Acinetobacter and determine carbapenem resistance in a few hours from a patient sample or positive blood culture. The first part of this is very doable using nucleic acid technologies. Verigene has a research-use test that ID's Acinetobacter spp, and other platforms surely have the same capacity. However, carbapenemase genes come in lots of varieties and developing a gene sequence test for them that has high sensitivity is no trivial task. And it's necessary, because 25-50% of Acinetobacter infections are still susceptible to carbapenemase, and you don't want to use the new, expensive antibiotic unless it is known that other agents won't work.

But let's say it's doable - what then? To have an impact, the hospital micro lab will have to test every Gram-negative blood culture with this test. Acinetobacter infections are on the rise, but they are still only a few percent of bloodstream infections in the US. Let's say they are 4%. That means that the micro lab will have to run - and pay for - 25 tests to catch a single positive case. Current PCR tests for MRSA, which are much simpler than a prospective carbapenemase test, cost $50-100. A dedicated CRAB test would have to cost at least $200, when fully capitalized and staffed. So the micro lab will have to spend $5000 to make a single diagnosis that drives use of the new $10,000 antibiotic.

That's a lot, but it still makes good economic sense, from a societal standpoint, to spend that money and treat the infection effectively. But society (at least in the US) doesn't make the spending decision - individual actors, such as doctors and microbiology lab managers do, and their incentives do not always align with society's.

We found this out the hard way at MicroPhage. Our test allowed patients with methicillin-susceptible S aureus infections (about half of all S aureus infections) to be taken off empiric vancomycin, and be treated with more-effective and less-toxic beta-lactams. It cost $50, and a typical hospital would have to spend about $2000 in testing to get an actionable result. The savings in health care costs would be $10,000+, making the test very cost-effective.

It was an utter flop in the marketplace, with sales of about $50K in its first and only year on the market. The problem? The micro lab manager, who had to buy the test, paid all of the costs and saw none of the savings. That was pretty much a deal-breaker.

Absent significant structural reform, the hypothetical CRAB test would likely meet the same fate. And without the test, the antibiotic would have much less impact, and likely fail to generate a decent ROI.

None of this is meant as a criticism of Spellberg and Rex. But it points out another layer of structural barriers that we have unwittingly erected to the development of new antibiotics. Unfortunately, until all of these problems are solved, it's as if none of them are solved.

 

Posted with Blogsy

The sore throat menace

The IDSA's report on infectious disease IVDs is out now, and it is squarely focused on the need for tests that will enhance antibiotic stewardship efforts. In particular, a test that would distinguish viral from bacterial infections is needed. Tens of millions of patients seek medical aid for upper respiratory infections each year in the US. About 10% of them have bacterial infections, but 60% or more go home with a prescription for antibiotics. Worse, the fraction of these prescriptions that are for broad-spectrum antibiotics is steadily increasing.

This is pure folly in so many different ways. Antibiotic susceptibility is a finite resource. Of the few sore throats that are bacterial infections, nearly all are due to Strep pyogenes, and nearly all strains are susceptible to penicillin. How many patients who went home with Cipro later ended up with C. difficile diarrhea or became infected by some cephalosporin-resistant bug?

A place where the report's authors did not go is to advocate restrictions on the ability of physicians to prescribe antibiotics as promiscuously as they have become accustomed to doing. Guidelines and education are all that is advocated. But this approach has been in place for decades and has stalled out: antibiotic prescriptions for upper respiratory infections went from 80% of patients in the 90's to 60% in the 00's, and has stayed there ever since.

 

From Barnett and Lindner 2013

Two developments are needed to have an impact. The technological fix would be a point of care test that distinguishes viral from bacterial infections, and this is certainly near the top of the IDSA's wish list. But availability of a test that will improve prescribing practices is not enough to ensure its adoption. We developed such a test at MicroPhage, and it sank like a rock in the clinical marketplace. There also needs to be some form of coercion, or if you like, encouragement, to do the right thing rather than the expedient thing. The growth of electronic medical record keeping means that it should be possible to track MDs who prescribe excessive amounts of antibiotics. A letter from the state medical board, or the FDA, might be a good way of gaining the attention of these miscreants. Or just knowing that someone is watching is often sufficient to improve behavior.

If this seems heavy-handed, consider that these physicians are creating a public health hazard, while providing minimal clinical benefit to their patients and exposing them to an increased risk of adverse events. I think that is sufficient rationale for impinging on physician autonomy.

 

Posted with Blogsy

Breaking good?

Scheduling time with a doctor is difficult now. If the Affordable Care Act survives and is successful, it is likely to get harder, as 10 or 20 million newly insured citizens will be added to the patient pool. Doctors will have to churn through patients even faster than they now do, or we will need more doctors.

Is this possible? Is the number of doctors limited by the number of people who have the desire and ability to qualify for an MD? Or are we limiting the supply of MDs by constraining med school admissions and certification of foreign-trained doctors? The answer is almost surely the latter. Western European countries have per capita about 3 doctors for every 2 that practice in the US. Perhaps it's just a coincidence, but they also have better outcomes and lower costs than the US as well. Although the US public does not seem to benefit from our doctor shortage, doctors do - they are paid much more than European doctors.

The ostensible reason for limiting the supply of doctors is to maintain the quality of licensed physicians and thus protect the public. But a look at the selection process for screening applicants to medical school suggests that other factors are just as important. .

I was reminded of this by a recent essay in the New York Times in which Barbara Moran, an aspiring med student, both deplores and defends the role of organic chemistry classes in weeding out prospective pre-meds. As anyone who has taken this class knows, pre-meds are frantic to get an A in it; anything less will disqualify them from admittance to med school.

The ostensible rationale for this requirement is that doctors need a grounding in O-chem to understand biochemistry and pharmacology. This may sound good to anyone who is ignorant of both fields, as well as of the practice of medicine, but it is complete bunk. Even if doctors did remember what they transiently learned in O chem many years prior, there is simply no medical scenario in which they would put their knowledge of the principles of alkene oxidation into practice.

It's true that doctors should have a good understanding of biochemistry. But human biochemistry involves a very limited number of types of reactions (plant and microbial biochem is another story), and these are all mediated by enzymes in aqueous solution at neutral pH, not by the various catalysts and solvents one learns about in O chem.

Moran acknowledges that O-chem is pretty free of any practical applications for the practice of medicine, but defends its inclusion on the pre-med obstacle course anyway. She quotes her teacher, who claims that organic chemistry teaches "...inductive generalization from specific cases to something you’ve never seen before." A useful skill in doctors (and other human beings) to be sure, but inductive generalization is a part of just about any intellectual discipline, from philosophy to astrophysics.

Knowing some chemistry is good. Chemistry explains the transformation of one form of matter to another, and the flow of energy that drives living beings. No one would want to be treated by a doctor who didn't understand acid-base or redox chemistry, or who didn't know what a protein is, or how vitamins work.

But getting an A in organic chemistry is not necessary for any of this, nor is it a guarantee that a prospective doctor really understands the fundamentals of biochemistry. What it does signify is that a student was willing and able to memorize long lists of formulas and recite them quickly on demand. This is not a useless skill - anatomy and lists of drug contraindications require good memory skills. But there are plenty of good ways to assess this ability - such as testing prospective doctors knowledge of anatomy and drug contraindications.

O chem is in reality a signifier, in much the same way that fraternity hazing rituals are. It shows a basal level of ability, but more importantly, it shows the commitment of the candidate to becoming part of the group. Or perhaps "caste" is the more appropriate term. Doctors, like other castes, are notoriously jealous of their prerogatives. They overwhelming oppose expanded responsibilities for nurse practitioners, despite 50 years of evidence that this does not result in reduced quality of care. They strive to maintain a dress code that sets them apart from their patients.

This is not the worst example of chemistry being used for a less than noble purpose. But it does make me wonder just how many potentially good doctors we have lost because they could not quickly regurgitate reaction diagrams that they would never use again.

 

Posted with Blogsy

A Tale of Two Diseases

Barry Werth's fine report in Technology Review describes not just the astonishing prices that new drugs can command, but explains how and when the high prices are justified - or not.

Kalydeco is the first effective treatment for symptoms of cystic fibrosis and is priced at $294,000/yr. Despite the massive cost, Kalydeco is being accepted by doctors and insurers because it really helps patients, and, perhaps, because only 2400 patients qualify for treatment.

Zaltrap is a me-too drug for treating metastatic colorectal cancer. It was priced at about $100,000 for a course of treatment - definitely on the high end for new cancer drugs, but not that different from Avastin.

For that kind of money, you'd expect pretty good performance. But Avastin and Zaltrap only increases survival for 6 weeks over standard chemotherapy, which is why Avastin has been rejected repeatedly for use in the UK for failing to meet the test of cost-effectiveness.

Despite its high cost and marginal performance, Avastin racked up nearly $6B in sales last year. That's because in the US, Medicare and Medicaid are required to pay for all FDA-cleared treatments. The FDA only considers safety and effectiveness, and gives no consideration as to whether a new drug provides a meaningful benefit over existing drugs.

FDA clearance is thus effectively a hunting license, allowing marketing departments to create a demand that overwhelms any objections based on cost. And when it comes to cancer, our willingness to pay has been unbounded. Paying $50-100K for an advanced chemotherapy regimen that extends survival by a couple of months is completely normal.

For antibiotics the economics are much different. Consider fidaxomicin, recently approved for treatment of C. difficile associated diarrhea. CDAD kills 14,000 Americans every year - not as many as colorectal cancer (50,000/year) - but no small number. Existing antibiotics, such as vancomycin, have a good global cure rate of some 66%, and fidaxomicin improves this to about 76%.

The cost? Vancomycin treatment is typically about $1200, and fidaxomicin raises this to $2800. In other words, for the cost of two weeks of cancer chemotherapy treatment you can get an antibiotic that will cure you of CDAD 75% of the time. And presumably not die.

The costs to develop a cancer chemo therapeutic are not markedly less than to develop an antibiotic. So is it any wonder that the cancer pipeline is full of drugs with marginal additional effectiveness? If you were the CEO of a pharma company, would you put your resources into developing a drug that you can charge $100K for, or one that you can only get $2500 for?

Antibiotics are simply too cheap for our own good.

 

Posted with Blogsy

On the other hand...

I've been pretty critical of hospitals' reluctance to make use of rapid diagnostics to improve patient care. And I think that this criticism has been well-deserved. But these tests do have some shortcomings, and it's only fair to take note of them.

Limited reportable results. Most rapid tests only identify a single organism. An exception is the Verigene Gram-positive Blood Culture Test, which returns ID results for 9 species. Limited results mean that labs still have to do a full workup in order to identify organisms not covered by the rapid test. Rapid testing therefore doesn't replace any test; instead it adds to workload and cost.

Resistance but not susceptibility. Rapid tests, with one exception, detect sequences or products from resistance genes such as mecA or vanA/B. A positive result thus (usually but not always) indicates a resistant phenotype. But when resistance rates are at 30% or 40% or even more, doctors tend to presume resistance and prescribe accordingly before receiving test results. Thus the clinical value of a positive result for resistance is somewhat limited - the patient is likely already on the antibiotic indicated by the test as being appropriate. And then, a negative result for a resistance gene is just that - it means a certain gene (actually just a small target segment of it) is missing or altered, but it does not necessarily follow that the organism will be susceptible to the antibiotic in question. Of the rapid tests, only the MicroPhage test was cleared to return a susceptibility result which would enable a change in therapy. And it is no longer available.

So there are some legitimate reasons to hold off on adoption of rapid testing. But they still are outweighed by the benefits to the patient, and to the healthcare system as a whole. You don't have to save many ICU days to pay for a whole lot of testing. But the savings are often diffuse and indirect and accrue to payors outside the hospital, while the costs are borne by the lab making the purchasing decision. Viewed from this perspective, it's not hard to see why adoption of rapid testing has been so slow.

Posted with Blogsy

What the CDC report didn't say

Judging from the number of Google alerts hitting my inbox, pretty much everyone who blogs in the healthcare space is writing about the CDC report on the threat of antibiotic resistance. Most of this writing is superficial regurgitation of the headline numbers, and is not really very interesting. The report itself is a nice resource that collects a lot of existing data in one place and makes it accessible to the general public, but does not contain any surprises or revelations.

The report rightfully calls for development of new antibiotics and rapid diagnostics to contain the threat. David Shlaes has posted a nice analysis of the short-sighted and flawed thinking at large pharma companies that holds back antibiotic development efforts.

A different, but no less crippling dynamic holds back the development of rapid diagnostics. As I've written before (here and here), the incentives for rapid diagnostics for antibiotic resistance and susceptibility determinations are misaligned, if not lacking altogether. Patients would benefit from their use, but no one in the hospital has a compelling reason to adopt these tests, which do tend to be pricey by the standards of microbiology. That is, they cost more than a Petri dish or a tube of rabbit plasma.

Hospital microbiology labs have to purchase the tests out of budgets that are always under pressure, given their status as cost centers rather than revenue generators. Micro labs in turn bear no responsibility when patients are prescribed inappropriate empirical antibiotics; they get no reward when an inappropriate prescription is avoided, even though this can result in considerable savings.

A well-respected clinical microbiologist once remarked to me that what MicroPhage really needed was for a patients' family to sue a hospital for not using our test, leading the patient to die while on ineffective antibiotics. He imagined the courtroom scene: "You mean that you left the patient on wrong antibiotics for 3 days because you didn't want to spend $50?" The thought of a penny pinching hospital admin squirming on the stand clearly intrigued him.

The unfortunate thing about this situation is that there is, I think, considerable willingness to pay to get it right. We just don't ask the right people - that is, the patients. I visited Shanghai in 2011 to get a sense of whether China might be a potential market for our test. Remarkably for a nominally socialist country, there is no system of socialized medicine, nor is insurance common. Patients pay out of pocket for care, which is one explanation for why their savings rate is so high. I asked ID docs at local hospitals to explain who would order our test and how would it be paid for. They replied that they would go to the patient or their family, tell them that there was a test that would help guide treatment, and ask if the family was willing to pay for it. They were pretty sure the answer would usually be yes, even at the US price of $50.

And I suspect that most families of American patients with S. aureus bacteremia, which has a mortality rate of some 30%, would give the same answer. But nobody is asking them.

Until that changes, or until hospitals start getting sued for inappropriate antibiotic use, the return on investment for developing rapid antibiotic susceptibility tests will remain questionable.

 

Posted with Blogsy

MDs playing lab tech?

David Shlaes is lamenting the passing of the good old days, when clinicians ran diagnostic tests right there in the ward. Gram stains, differential culture - the whole shebang. Even collected specimens themselves, they did. I'm sure this was great fun, and there is certainly value in having MDs get a little hands-on experience, if only to gain an appreciation of what can go wrong.

David is a smart guy, and knows way more about antibiotics than I will ever hope to - but this is plainly silly. There may be some overlap between doctoring skills and lab skills, but it is pretty slight, in my experience. There are very few MDs (usually MD-PhDs) I've met that I would want to have working in my lab. Not because they weren't smart enough, but because they weren't skeptical enough of their own genius to repeat "interesting" results over and over to see if they were real.

There's a reason why diagnostic tests are run by dedicated personnel who have to demonstrate proficiency and document their methods and results thoroughly. It's not to maintain full employment of lab techs, but to ensure reliability.

But his larger point is very real: the benefit of rapid diagnostic tests is pretty much lost when it takes half a day to get the specimen to the lab, and another half day to report the results. Real-world turnaround times for 2 hour PCR tests are typically 12-18 hours, pretty much destroying their usefulness.

The solution to this lag is not to put a GeneXpert on every floor and let MDs play with them. It's to stop treating microbiology labs like the red-headed stepchildren of the hospital family. Few microbiology tests are reimburseable; the hospital has to pay for them, and thus administrators see micro labs as cost centers to be squeezed, or better, eliminated.

The results are predictable - half of all S. aureus bacteremia patients are on the wrong antibiotic, because MDs are forced to guess at the appropriate treatment absent timely lab results. As more bugs become resistant to more antibiotics, this situation will only get worse.

The rapid (technically, accelerated) MRSA/MSSA test that my team developed at MicroPhage was a commercial failure. This was largely because there was no constituency in hospitals to pay for faster susceptibility results. The healthcare system as a whole would have saved substantial amounts of money as a result of fewer hospital and ICU days incurred by appropriately treated patients. But additional testing means additional lab costs, and few lab managers were interested.

The economic incentives to develop rapid tests are still all wrong. Until that gets fixed, patients will still suffer and die from inadequate treatment, no matter how many doctors we let play lab tech.

 

Posted with Blogsy

And I'm back...

I've been neglecting this blog in favor of business, family and otherwise, but mostly to enjoy the Colorado summer before it is gone.

One of the trips I took was to Trapper's Lake, one of the largest in the state, and also one of the most remote and scenic. At least it was scenic in Aug of 2002, the last time I was there. 2002 was one of the first big drought and wildfire summers in our new climate regime. Several nasty fires had sprung up already, and the sky in Boulder was relentlessly scorched and smoky all summer.

We were hoping to find some relief in the high country, and planned a family backpack trip looping south from Trappers into the Flattops Wilderness, a rolling plateau of lakes, forests and meadows. At 11,000 feet, we figured it would be high enough to be cool and lush despite the drought.

Unfortunately a fire had sprung up near Big Fish Lake, a valley just a few miles west of where we planned to hike. However, a call to the district ranger reassured us that recent cooler weather had calmed the fire down to the point where it was not a concern, and we proceeded as planned.

The hike was everything we had hoped for - cool temperatures, green forests and meadows, and limitless views from the plateau:

By our third day out, however, we began worrying about the smoke on the horizon, and thought it best to start heading back to the trailhead:

We camped at Parvin Lake, and the next morning it was clear that the fire had blown up. Worse, there is only one road out from Trappers, and the fire was heading toward it, possibly cutting us off.

When we got down on the valley floor by the lake, smoke from the fire was rolling over the ridge, turning the sky a dark, evil-looking red. Even more disquieting was the low rumble of the fire - it had become a monster that was rolling through the forest.

We split up at the lake, my wife and daughters taking the short route to the lodge around the east side. I took the longer trail along the west side where our car was parked. Short cutting the trail through a boggy area, I came upon a middle aged woman and her elderly mother, lost, terrified, and heading the wrong way. I walked them (all too slowly) back to the trailhead, jumped in the car and hightailed it down to the lodge. There I found a ranger screaming at my wife to "Get in the truck, now!" and leave with him before the road out was cut off. We wasted no further time with him and started driving as fast as we could down the rough dirt road.

We rounded the ridge and saw flames shooting hundreds of feet in the air, as if the atmosphere itself was on fire. The trees were dwarfed by the size of the flames, which had become a red mountain piled on top of the green mountains. I took a very quick picture:

The trip this year was quite a bit less exciting. And although the frosts are mostly gone - they will take decades to regrow - Trappers Lake is still a place of tranquility and beauty

 

 

Posted with Blogsy

Combo therapy fails

A common question I've heard when discussing the case for accelerated antibiotic susceptibility testing is "why not just use several drugs in combination?", the presumption being that one of them is sure to work. The usual answers are that there is a risk of toxicity, and that pharma companies are generally not interested in sponsoring combo therapy trials - they want to own markets, not share them.

Two articles published in CID (thx @ndm1bacteria) add a third and more compelling reason: combo therapy does not work nearly as well as one would expect. A study of Acinetobacter infections treated with colistin and rifampin, and a study of Pseudomonas infections treated with various antibiotics, showed no improvement in 30-day mortality for combination over single therapy.

This is a surprise, given the success of retroviral combination therapy, and the results of in vitro microbiological studies. Why the lack of efficacy? I can think of three factors: pharmacokinetics and biodistribution, bacterial SOS responses, and host inflammatory response. It's possible that a significant fraction of the infecting bacteria did not (at least initially) get exposed to a significant dose of both antibiotics. And when exposed to one antibiotic, bacteria (unlike retroviruses) will go into stress/shutdown modes that make them less susceptible to additional antibiotics. If the infection is not rapidly eradicated, the host inflammatory response can spiral out of control, leading to organ damage and death.

In contrast to combination therapy, appropriate antibiotic therapy was shown to improve survival. The bottom line here is not surprising: using the right antibiotic matters, and a shotgun approach to therapy is not a substitute for evidence-based medicine. The problem is with the evidence part: accelerated antibiotic susceptibility tests still don't exist. And they will continue to not exist until physicians begin to demand them and lab managers show a willingness to pay for them.

 

Posted with Blogsy

Mass spec - not the droid you are looking for

Maas spectrometry exerts a peculiar hold over the minds of its advocates. They tend to be passionately committed to it, like no other technology I've seen.

MS is indeed a powerful technology. A sample is ionized so that it has a net electrical charge, and the time that it takes to fly toward a cathode is used to calculate its mass (more time = more mass). The identity of compounds in a sample can be deduced from matching expected with observed molecular weights. Under ideal conditions, only a few femtograms of material are needed. Good stuff indeed.

When complex samples, such as tissues, cells or fluids are analyzed, a complex pattern of peaks is observed. These patterns are very data-rich, and can be mined to produce a signature, and the signatures can be used for clinical tasks such as predicting therapeutic response to cancer drugs - or identifying bacteria.

The allure of this technology to a micro lab manager is obvious - a universal platform for identifying bacteria. A loopful of bacteria from a colony are smeared on a plate, overlaid with an acidic solution, zapped by a laser, and out comes a readout that IDs the sample.

When it works (and it works most of the time) it is beautiful and very rewarding. But it doesn't work all of the time - some species are just not distinctive enough in their signature to give a good result. More importantly, MS is very sensitive to the sample matrix, meaning that it doesn't work nearly so well on direct specimens. As a result reliable ID will still take at least a day, in order for purified cultures to be prepared.

I think this very good but not perfect success rate is the key to the enthusiasm (if not obsession) that many mass spec'ers exhibit. When you train a dog, you do not always give the treat - the uncertainty in the dog's mind as to whether she will be rewarded causes her to pay much closer attention to her master. Casinos operate on the same principle - provide just enough reward to create hope and anxiety, and the reward becomes much more gratifying and addictive. The occasional failure serves to redouble the true believers determination to succeed.

I was glad to see that the recent paper by Donna Wolk and colleagues takes a level-headed approach to the strengths and weaknesses of mass spec, despite its title. And it is likely true that mass spec will instigate a "fundamental shift in the routine practice of clinical microbiology". I just don't think patients will see much benefit.

Although all clinical microbiologists are trained to focus on organism ID, this parameter is decreasingly important in an age of widespread antibiotic resistance. An instrument that can ID hundreds of potential pathogens is nice for lab workflow but consider: just 4 organisms account for 80% of positive blood cultures (counting Coagulase-negative Staph as one organism); 6 account for 90%; and 7 for 95%. The ability to ID a Citrobacter or Proteus will help very few patients. Knowing whether a septic patients' Gram-negative rod is susceptible to cephalosporins or carbapenems has much more therapeutic utility, regardless of whether it is an E. coli or a Klebsiella.

There have been some initial efforts to determine susceptibility and resistance using MS - looking for specific enzymes, or antibiotic degradation products, or even cellular response to antibiotic exposure. I think all of these will ultimately fail because of the vast heterogeneity of resistance mechanisms and differences in strain responses to antibiotic challenge. Even with copious data storage and analysis capabilities, the signature of each strain of each species to each antibiotic over a range of physiological conditions will have to be validated, and have to be constantly updated as new strains and resistance mechanisms emerge. This is implausible.

What would really benefit patients (as opposed to lab managers) is a same-day test that would return ID results for the top 4-6 pathogens, with susceptibility results for 2-3 frontline antibiotics each. Such a test would ensure that 90% of critically ill patients receive appropriate antibiotics, instead of the 50% that now do.

But that brings us back to the question of the economics of test development. We were working on a panel test at MicroPhage. But with microbiology labs unwilling to pay $50 for our test that returned same-day results for MRSA/MSSA, how much would they have been willing to pay for a small panel? Potential partners and investors surmised that the answer was "not enough", and the project never really got traction. So long as antibiotics and susceptibility testing are undervalued, the prospects for developing a test that will get patients the right antibiotic right away are not good. Instead what we will have are $150,000 instruments like mass spectrometers that improve workflow and reduce staffing costs.

 

Posted with Blogsy

Device tax hysteria

There is something about taxes - also known as paying for a civilized society - that just drives a certain segment of the populace crazy. There is no better example than the medical device tax, enacted as a funding mechanism for Obamacare.

The tax is set at 2.3% of sales of medical devices - pacemakers, stethoscopes and the like. Is this unfair, or worse, counterproductive? Well, the device makers certainly think so. "Medical technology companies across the country are struggling to remain competitive..." says Steven Ubi, the president of the industry group Advamed (I used to be a member). According to Senator Dan Coats, one company can no longer afford to continue "...working on solutions to relieve a wheelchair-bound child’s discomfort" because of the tax. Oh my, Obama is going after crippled children now. The Wall Street Journal says the tax is "...bound to destroy startups and stunt medical-device innovation in the U.S. and thus the quality of health care world-wide." So it's not just crippled children, it's the whole world that will suffer.

Really? A 2.3% tax is going to bring a flourishing industry to its knees? Someone should tell investors. The iShares Medical Device Index (IHI) has posted 14% annualized returns in the three years since Obamacare was passed. Since the tax was implemented at the start of this year, returns are 16%.

It's not too hard to figure out why investors, as opposed to self-dealers and gasbags, are not worried. Gross profit margins (sales receipts minus production costs) are typically 60-70% in this industry. Hospitals, who are the primary customers, typically markup this sales price another 100% or more when billing patients and insurers. 2.3% is not the end of an industry. It's not a crippling death blow. It's just noise.

About 50 million Americans lack health insurance, 16% of the population. If Obamacare cuts this number in half, medical device manufacturers will see a market increase three times as large as the device tax. That's why their stocks keep going up - they are about to get tens of millions of new customers. You'd think the manufacturers would be, if not actually grateful, at least discreetly content. But no. They want the new customers, sure. But they don't want any part of paying to bring them into the system.

This is truly swinish behavior.

 

Posted with Blogsy

What FMT can learn from PT

Fecal Microbiota Transplantation (FMT) may emerge as a serious alternative to antibiotic therapy. Phage therapy has not, despite many more decades of research and clinical experience. It's not as though the potential of phage therapy is a secret - there are any number of editorials and reviews that propose we give phage therapy "another look", as we combat increasing rates of antibiotic resistance. Yet there are only 4 phage therapy trials listed on ClinicalTrials.gov, and one of these appears to never have gotten started. By contrast, 18 are listed for FMT therapy, a number that is surely an undercount given how recently the FDA began insisting on formal trial protocols.

Phage therapy has been around for many decades, predating the antibiotic era, and continuing in use in Eastern Europe. Although only one placebo-controlled clinical trial has been reported, it gave encouraging results for treatment of otitis media due to Pseudomonas infection. But those results were published in 2009. The sponsor, BioControl (now AmpliPhi Biosciences) no longer features the otitis project on the company's website, although there is a respiratory Pseudomonas project in preclinical development.

I would suggest that phage therapy has an insuperable intellectual property problem. Phage aren't like other drugs. It's not just that they are alive (or biological, if you prefer), it's they are too easy to make and too hard to define.

Phage experiments are really easy to do. When physicists with no lab skills decided to "solve" biology in the mid 20th century, they turned to phage as model systems. Phage are everywhere in the biosphere, they multiply by many orders of magnitude in a few hours, and they are conveniently stored in the fridge when it's time to call it a day in the lab. Because they are so prolific and so easy to handle it's easy to select phage that have desirable properties, such as enhanced killing of pathogenic bacteria.

You can obtain patents on these phage, but it won't protect your company from completion. We never filed on our proprietary phage at MicroPhage, because it would just be too easy for someone else to independently select phage that were functionally equivalent yet compositionally different, and so did not infringe.

This is why no pharma company or other deep-pocketed entity has ever put serious money into phage therapy. Imagine that you developed a phage cocktail that suppressed wound infection. There's little to stop a second (or third, or fourth etc) group from doing the same thing, as the concept of phage therapy of wounds is in the public domain. It would be essentially impossible to show that one formulation was superior to another in clinical trials. That would leave your company to compete on cost, driving everyone's margins down to nothing. Generics compete on cost, but the makers of generics don't have to spend hundreds of millions of dollars to perform clinical trials before they see a cent of revenue. No one is going to put up the money for high-risk clinical trials whose best outcome is a product with a small addressable market and low margins.

This is a failure of our profit-driven system (to be clear, I have enthusiastically participated in and profited from this system). There is no doubt that patients would benefit from phage therapy. It is becoming increasingly clear that patients are benefitting from FMT. A key question for the future of FMT is whether it can avoid the same IP trap as phage therapy. I don't think so. Like phage cocktails, I expect that FMT cocktails can be diverse in composition, yet functionally equivalent. There are dozens, if not hundreds of bacterial species to choose from in making a defined FMT cocktail, and an unlimited number of strains within species. It is telling that of the 18 FMT trials listed on ClinicalTrials.gov, not one is sponsored by a commercial entity.

Thus there is a serious chance that FMT will end up like PT - a marginal or experimental therapy practiced by a few experts. The only way forward that I see is for a not-for-profit (perhaps crowd sourced?) entity to pick up the costs of developing a manufacturable formulation and putting it through clinical trials. The standard model of venture or pharma financing is just not going to work.

Posted with Blogsy

Antibiotics are too cheap

It's a common trope to bemoan the lack of resources that pharma companies put into antibiotic development. But why should they, when so much more money can be made in other fields? The newest example can be found in the excitement over Xofigo, which increases median survival in metastatic prostate cancer from 11 to 15 months. A positive step, but calling it a "big deal", as the Cleveland Clinic's Robert Dreicer did, speaks more to how awful most cancer treatments are.

The cost? $69,000, fully reimburseable by insurance. And the side effects? 47% serious adverse events vs 60% for the standard treatments. This may be the best thing out there, but it is still a terrible drug by any objective standard.

Compare this to the performance of anti-MRSA drugs. Without antibiotics (or inappropriate ones) the 30-day mortality for bacteremia is > 80%; with appropriate antibiotic treatment it is about 25%. Serious adverse event rates are typically a few percent. Drug costs are some $120 for vancomycin, and 10 times that for daptomycin, which is considered to be scandalously expensive.

It would be nice if pharma companies were willing to spend hundreds of millions of dollars to develop a drug that they can sell one time for a hundred dollars; but we don't live in that world. The fault is ours: we fear cancer and are willing to spend almost anything to hang on, sick and miserable, for a few additional months. We don't fear bacterial infections nearly enough, it seems.

Antibiotics are too cheap. As Laurie Piddock put it: "The price of antibiotics needs to relate to their value to society and should not relate to the price of previous products." Until we begin to value a life lost to infection as highly as a life lost to cancer we can expect to see little investment in new ant-infectives.

 

Posted with Blogsy

Rapid diagnostics - ID is not enough

In a previous post I discussed some of the hurdles - mainly economic - to developing rapid diagnostics for bacterial infectious disease. Despite these problems, several new technologies have been developed and are finding their way, however slowly, into clinical microbiology labs.

The problem with all of them is that they are great at identifying bacteria, but are limited in their abilities to determine resistance and susceptibility. In a low-resistance environment, this is not a significant limitation. Historically, doctors would consult a reference that matches clinical presentation and bacterial species with preferred antibiotics, and prescribe accordingly. They might also consult their hospitals' antibiogram, a tally of observed antibiotic susceptibilities at their hospital.

But what do you do when the antibiogram looks like the one below (which is very typical)?

 

Your rapid test may have identified the infecting bug as Acinetobacter, but then what? None of the antibiotics on the list are highly likely to be effective for any particular strain, but there is a pretty good chance that one or two might be. You just don't know without a susceptibility report on the bug that has been isolated from the patient - and that still takes 2-3 days. The rapid ID test has told you what the patient is infected with, but not what treatments are likely to work.

This limitation is true whether the rapid-test technology is PCR, PNA-FISH, or the new favorite of many, mass spectrometry. It's true that PCR tests can detect the presence of some resistance genes. But the prevalence of these genes (e.g., mecA in S. aureus, vanA in E. faecium) is so high that doctors usually assume resistance and treat accordingly regardless of the test result. A negative result for a resistance gene does not necessarily mean that the strain is susceptible, as there can be new variants and new pathways for resistance that gene tests may not detect. Accordingly, no gene tests have been cleared by the FDA for determination of antibiotic susceptibility. The distinction between negative and susceptible is not always obvious to doctors, and the FDA has required the Cepheid MRSA PCR test to carry a label to that effect.

The basic problem is that antibiotic susceptibility and resistance are complex phenotypes that can't be reduced to a problem of detecting a few molecules or gene sequences. Resistance can occur through thickening or modification of the cell wall, chemical modification or mutation of ribosomes, or production of degradative enzymes and efflux pumps. Each of these mechanisms can have many variants and pathways, and new ones are always evolving. Even if (actually, when) it becomes possible to get an entire genome sequenced within a few hours, it will not be possible to read out a susceptible phenotype with high confidence. There are just too many possible genetic variants that would have to be validated first.

So for now and the indefinite future, ID-only rapid tests will continue to have limited clinical impact, and susceptibility testing will have to proceed by traditional methods: expose a bacterial strain to the antibiotics of interest, and observe the response. This process typically takes 2-3 days at present. Some methods to reduce this time are being developed, and I will write about them in a later post.

 

Posted with Blogsy

Rapid Diagnostics

In previous posts I wrote about the three principal barriers to new antibiotic development: economics, biology and regulatory approval.

I'll add another. A really great diagnostic would be of immense help in executing clinical trials. Enrollment in trials is often a problem. Patients may be initially treated with an antibiotic other than the test or control antibiotic. The infecting bug may not be the one indicated for the new antibiotic. Or it may be equally susceptible to both the control and test antibiotic. All of these issues increase the cost and decrease the power of a clinical study.

A rapid diagnostic that identified the infecting bacteria and determined antibiotic resistance and susceptibility would go a long way to resolving these issues. Right now, standard methods require 2-3 days to return results, too late to be of use in trial enrollment.

What would a great bacterial infectious disease diagnostic device look like? As always, the ideal device would be perfectly accurate, give instant results with no specimen workup, and cost nothing. That's the fantasy anyway.

Having tried, succeeded (in getting regulatory clearance) and failed (in the marketplace) in developing such a device, I have a few thoughts about the challenges involved.

First, as always in our system, is economics. No one is going to develop a diagnostic device solely for the purpose of enabling clinical studies. There has to be a viable market for the device in routine clinical use. This requirement turns out to be a surprisingly severe constraint.

There are lots of visits to doctors offices and ERs for infectious disease complaints. But the vast majority of these are for non-lethal indications: urinary tract infections, wounds, sinusitis, otitis, bronchitis. At MicroPhage we looked long and hard at the case for developing a urinary tract infection (UTI) test. There are many millions of prescriptions written for these every year, and the rate of resistance of the principal pathogen (E. coli) to frontline antibiotics is high, 10-30%.

Sounds like a great business opportunity, right? Big market, high rates of resistance, lots of ineffective scrips being written. But what is the consequence of getting the antibiotic wrong? The patient will experience a couple additional days of discomfort, then call her doctor and get a different prescription. That's it.

How much would she (or her insurer) be willing to pay for a test to avoid this inconvenience? How long would she be willing to wait in the doctors office while a tech ran the test? After all, antibiotics are cheap and her time is valuable. The principal harm done is to further weaken the effectiveness of frontline antibiotics such as ciprofloxacin. That is a cost that is diffused throughout society, and no one is willing to pay to prevent it.

The story is much the same with wounds and most other infectious disease indications. There just really aren't that many indications where rapid identification and susceptibility results are critical to the patient's well-being. Hospital-acquired/ventilator-associated pneumonia is one. But that market is tiny and getting samples is hard.

Bacteremia is another critical indication. There are about 5 million positive blood cultures in the US per year (several per patient on average). About 100,000 patients will die. There is good evidence that getting the right antibiotics to the patient reduces length of stay and mortality for S. aureus bacteremia. But only about a million of those positive blood cultures are Gram-positive cocci in clusters, and are thus potential S. aureus bacteremias.

Venture capitalists simply don't get excited (greedy would be the less kind term) about market sizes less than a billion dollars. Even if you could charge $100 for your MRSA/MSSA test, and owned the entire market, you would fall well short of arousing the interest of serious money for developing a new technology.

And rest assured, few microbiology labs will spend anything like $100 per test. PCR MRSA tests, like Cepheid GeneXpert and BD GeneOhm, cost $25-75 on top of a capital outlay of $25-150K. They have been adopted by only a few percent of hospital micro labs. The MicroPhage MRSA/MSSA test was $50 and was adopted by about a dozen labs in the year that it was on the market. The attitude at Northwestern University Hospital, one of our clinical trial sites, was typical: "It's a nice test. I would buy it if it was $5."

That's unfortunate. Several very nice studies (Brown, Goff, Schweizer) show that getting the right antibiotics to S. aureus bacteremia patients in a timely fashion decreases mortality and length of stay in both the ICU and the hospital. Cutting just 1 or 2 days stay in the ICU pays for an awful lot of testing, even at $50 or $100 a pop.

So why are hospitals so unwilling to spend this money? After all, most bacteremias are hospital acquired, and so those extra costs (and deaths) are supposed to be the hospitals' responsibility. Even under the most conservative assumptions, rapid testing should be a great investment for hospitals.

The problem is that the costs are all borne by the micro lab, whose budget pays for the test and the staff to run it. The savings accrue to the hospital as a whole - no one gets to take credit for them, and thus no one is incentivized to champion rapid testing. Furthermore, the hospital has a strong incentive to game the system. If they attribute the extra care to causes other than the infection acquired in the hospital, then they can get fully reimbursed by insurers. Gaming ICD codes is much easier than implementing a new testing and reporting scheme. Everybody wins - if we define "everybody" as the hospital and micro lab. The patients and insurers not so much.

The bottom line here is that the market for rapid ID/AST is not great - markets are small and hospitals are reluctant customers at best. So long as antibiotics and antibiotic stewardship are not highly valorized there is little incentive to invest in the new technology needed to facilitate antibiotic clinical trial execution. There are also significant technical barriers, particularly with respect to prospects for a rapid AST, and I will discuss these in a future post.

 

Posted with Blogsy

Glen Smith, 1929-2013

My father passed away at home on Friday July 5. He died at peace, without pain and surrounded by family. He always had a horror of institutionalization, and I am grateful that I was able to help him have a good death. I am especially grateful to Sangre de Cristo Hospice, who managed his pain and supported my mother with compassion and competence.

Both birth and death are over-medicalized in our country. You have to be assertive if you want these events to be a part of life, rather than become a source of medical procedures to be performed. My Dad's prostate cancer had been in remission for 16 years. When the source of his back pain was identified as bone metastases, he tried to make it clear to his oncologist that he did not want aggressive therapy, but only to die without suffering and without making his family suffer.

The oncologist talked him into a course of Lupron therapy (a testosterone antagonist), even though his testosterone levels were negligible. Lupron has fairly mild side effects compared to cytotoxic therapy. One of these is bone loss. So in addition to Lupron therapy, another procedure is added (phosphonate therapy), to minimize bone weakness. Phosphonates act as T-cell activators and often cause flu-like symptoms. My dad then spent 3 of his last 40 days in bed with fever, aches and nausea in order to mitigate the side effects of a therapy that had no chance of improving his life. Because that is the procedure. And performing procedures is what doctors do.

Nor did the oncologist consider the impact of asking a sick 84 year old man with severe back pain and COPD to come in to his office once a week or so. He seemed genuinely surprised to learn that this constituted any sort of hardship. It was only with reluctance that he agreed that hospice care was the best choice for my father.

I was not asking his permission, as I have spent too much time around doctors to be overly impressed by them. However for many patients, especially of my parents generation, acquiescence with no questions asked is the norm. This may be the most convenient arrangement for doctors, but it is essentially BS, and the sooner it changes the better off we all will be.

Posted with Blogsy

WGS in the micro lab

Nucleic acid technologies are making steady inroads into the clinical microbiology lab as they get faster, better and cheaper. The logical endpoint of this development is the use of microbial genome sequencing as a diagnostic method. Given the pace of improvement in sequencing technology, the feasibility of routine sequencing of clinical isolates is inevitable. The question is whether this technology will add any value.

Writing in the Journal of Antimicrobial Chemotherapy, Torok and Peacock say the answer is yes: "...we believe that rapid whole bacterial genome sequencing has the potential to transform diagnostic clinical and public health microbiology in the not too distant future."

I am more skeptical and think the value of whole genome sequencing will be limited:

ID/Speciation: Genome sequencing is overkill. All the DNA information needed to speciate an isolate is contained in its ribosomal RNA genes. PCR and oligonucleotide hybridization methods return gold standard speciation results now. Additional sequence information will just be noise.

Epidemiology: Probably the best use of WGS information, as the additional sequence data allow the spread and evolution of strains to be tracked. This has already happened in the case of the KPC outbreak at NIH, and will become more common.

Resistance and susceptibility testing: A seductive but terrible idea. The response of bacteria to antibiotics is a complex phenotype, involving many genes. Reliably predicting this response would require a thorough understanding of the actions of all these genes. Even more difficult, it would require us to be able to predict the effect of various mutations in these genes on their activity. For example, a point mutation that changes the amino acid sequence of a metallo beta lactamase could make it more active, less active, or have no effect at all. There are billions of possible mutations in hundreds of genes that would have to be accounted for in order to reliably predict antibiotic response. This is not going to happen soon, or probably ever. More to the point, what actionable information would WGS provide that phenotypic susceptibility testing does not?

Don't get me wrong, I love this technology. I wish it had been around when I was in grad school - synthesizing oligos manually and then sequencing them by the Maxam-Gilbert method was some of the most tedious lab work I've ever done. But the application of WGS to clinical microbiology is likely to be much more hype than substance. Even the epidemiological applications could potentially be done just as well by much more old-fashioned methods such as phage typing (if anyone still knew how to do this). Just because it's new doesn't mean it's better.

 

Posted with Blogsy

Why there aren't more antibiotics, part 3

Lack of financial incentives, and problematic biology are two big reasons for the lack of new antibiotics. The third is the difficulty of gaining regulatory approval.

To be marketed in the US, drugs must be cleared by the FDA, which requires proof that they are safe and effective. The highest-quality evidence of safety and efficacy comes from randomized and blinded placebo-controlled clinical trials: patients are randomly assigned to be treated with the new agent or a placebo. Neither the patient nor the caregivers know who is given a placebo and who is given the drug - this insures that all aspects of treatment and evaluation are unbiased. Trials of new antibiotics are never done this way, for reasons I'll explain below.

Doing clinical trials has an inherent ethical danger - they are essentially experiments on human beings. The Nuremburg code was developed after World War II in response to the horrors of human experimentation carried out by the Nazis. One of the principles of the code is that there must be informed consent. Another, one that is particularly relevant to the conduct of antibiotic trials, is that the participants must have some expectation of therapeutic benefit commensurate with risk. Both of these principles were violated by the notorious Tuskeegee syphilis experiment. Effective treatment was withheld from the subjects without their knowledge, so that the natural course of the disease could be better understood.

When no treatment exists, or if it is relatively ineffective, clinical trial design is fairly straightforward. One group of patients is given the current ineffective therapy, and the other is given the new therapy. Demonstrating the superiority of the new treatment in these circumstances is quite feasible.

Antibiotics have historically been very effective, curing the large majority of patients who have bacterial infections. Thus withholding antibiotic therapy from one group of patients (as in the Tuskeegee experiments) is simply out of the question. The control group in an antibiotic clinical trial typically receives an antibiotic that is expected to be effective against the infection. In this type of trial, the most that the new treatment can hope to show is that it is not inferior to the control treatment. This is not the sort of conclusion that gets your new drug breathless write-ups about being a breakthrough - the kind of publicity that new anti-cancer therapies that extend survival by a few weeks or months often get. This publicity matters because it affects our willingness to pay premium prices, and thus create financial incentives.

Noninferiority is a complex statistical concept with a somewhat arbitrary definition, and it often relies on historical comparisons to trials that may have been performed decades earlier. And again, what you'd really like from a newly approved drug is a demonstration that it's better than the current standard of care. Otherwise, why bother to switch?

This aspiration, among others, has led the FDA to ask for demonstrations of superiority as a condition for clearance, even when demonstrating superiority was not really feasible. Replidyne's application to clear faropenem for bronchitis, pneumonia and wound infections was deemed not sufficient for approval, despite data from 11 clinical trials that enrolled 5000 patients. These trials were designed to show non-inferiority; by the time they were done the FDA decided it wanted superiority. Replidyne did not survive this raising of the bar and eventually folded, taking many millions of investors dollars with it.

Given the difficulty of finding new antibiotics, the relatively low return on investment, and the considerable regulatory risk, it is no surprise that many pharma companies have terminated their antibiotic development programs.

The Infectious Disease Society of America is particularly troubled by regulatory roadblocks, and particularly determined to to provide alternative pathways to clear drug candidates. The IDSA is pushing for Limited Population Antibacterial Drug (LPAD) labeling, which would require less extensive clinical validation, and be restricted to seriously ill patients with few therapeutic options. In addition, IDSA has published a white paper proposing improved and feasible clinical trial designs.

Both of these approaches would benefit greatly from the availability of rapid diagnostics that could determine antibiotic resistance and susceptibility in a few hours. These companion diagnostics would identify patients who are at risk of treatment failure for first-line antibiotics, and thus are most likely to see a benefit from new drug entities. More about that in future posts.

Posted with Blogsy

Hype Watch July 1

The Facts: A nanoscale cantilever device can detect differential responses of bound bacteria to antibiotics through changes in oscillation frequency.

The Hype: "Thanks to this method, it is quick and easy to determine if a bacteria has been effectively treated by an antibiotic.... Easily used in clinics..." Science Daily

The Reality: Development of rapid diagnostics for ID/AST (identification/antibiotic susceptibility testing) is not being held back by a lack of awesome biosensor technologies. Those are as common as warts on a witch. It's the pre-analytic steps that are the problem. The indications for which rapid ID/AST are really needed either have vanishingly low levels of bacteria (bloodstream infections) or samples in which pathogens are mixed with commensals (pneumonia).

The nano-cantilever offers no obvious innovation in bacterial ID. Although it detected a signal from just a few hundred bacteria, the sample applied held 10e5 cells, presumably at a high concentration, 10e7/ml or more. Since most positive bacteremia samples have less than 1 bacterial cell per mL, there is a 7 log gap in current vs needed sensitivity.

The verdict: if the pre-analytic problem can be solved and a method of rapid ID can be coupled to it, then nano-cantilevers could potentially give AST results in minutes, rather than hours or days. That's a pretty big "if".

 

Posted with Blogsy

Spreading the wealth

One of the underappreciated side effects of antibiotic use is its ability to promote antibiotic resistance. It's no surprise that antibiotics would selectively enrich resistant strains of bacteria, favoring their reproduction over that of susceptible strains. More unexpected are the findings that antibiotics induce a shotgun blast of resistance genes into their local environment, where they may find new homes.

Bacterial genomes play host to bacterial viruses (bacteriophage) that quietly reproduce along with the host, and sometimes encode genes that are of some benefit to their hosts - including genes for antibiotic resistance. However, when these viruses sense that the host is in trouble, they will excise themselves, replicate (usually to the point of killing the host) and leave to find a new home.

Evidence is mounting that this phenomenon - which has been known for 70 years - can cause the spread of resistance both within and between different types of bacteria, including MRSA, Enterococcus and the gut microbiome in general. The upshot is that using an antibiotic increases the likelihood that previously susceptible strains will be made resistant, chipping away at the usefulness of that antibiotic.

The odd thing about this behavior is that it spreads the resistance gene at the expense of the individual bacterial host - despite harboring the resistance gene, the host is killed anyway by its bacteriophage parasite. This most likely occurs because the bacteriophage jump ship whenever they sense that the host is in distress, whether the source of the stress is starvation, excessive heat, toxic chemicals or antibiotics.

This behavior may be an example of selfish genes in action. If antibiotics are in the environment, the selective advantage of a resistance gene is at a maximum - it is a sellers market for them. In killing their original host, they have made tens or hundreds of copies of themselves. If more than a few percent of the infectious particles find new homes, then their strategy has been a success, and will continue to be successful as their newly-resistant hosts outcompete susceptible rivals.

However, this rationale is true regardless of whether the host is stressed or not - if by killing the host the phage progeny can find 5 new hosts to infect, then obviously it is to their advantage to do so. Lytic phage follow precisely this strategy, killing every host they infect in order to reproduce as quickly as possible. Not surprisingly, bacteria have evolved a number of mechanisms, such as CRISPR and modification-restriction systems to thwart bacteriophage infection.

Thus the selfish gene view of antibiotic-induced resistance gene mobilization needs a corollary: that antibiotic-induced stress renders bacteria more susceptible to phage infection, possibly by suppressing bacterial defense mechanisms.

A study from James Collins lab provides indirect evidence that this is the case. Modi et al exposed mice to ampicillin and ciprofloxacin, and followed the spread of phage-encoded resistance genes. Resistance genes not only were more abundant after antibiotic exposure, but they were linked to a more-diverse set of bacterial genomes. In other words, it appears that bacteriophage were able to infect a wider variety of bacterial strains and species after antibiotic treatment than they were before. These findings support the idea, that from the resistance genes' selfish point of view, it's a good idea to look for new hosts when antibiotics are around.

The upshot is that antibiotic use not only encourages the spread of resistance by differential survival of resistant bacteria, but by dissemination of resistance genes into the environment. These genes can be picked up by different species that may not previously have served as hosts. This mechanism suggests a new danger of imprudent use of antibiotics: it allows resistance genes found in nonpathogenic strains to find their way into pathogens.

Posted with Blogsy

More on looking the part

Mike Edmonds has a few comments regarding the study on lab coats and cleanliness from Dr. Silvia Munoz-Price and colleagues.  Lab coats are washed about every 12 days on average, even though 90% of the respondents were aware that they were potentially contaminated.  The most prevalent reason given for wearing lab coats? (hint, it is not to do lab work) - Instead, it is "to symbolize their profession", a reason that must be immensely gratifying to some.