[Ed note: Following is another installment from our favorite medical scribe, “Doc Gumshoe” (no, he’s not a doctor). We feature Michael’s commentary every couple weeks or so, and his thoughts and words are his own. “Doc” doesn’t pick stocks, but FDA decisions are a huge driver of investment success for pharmaceutical and biotech companies and today Michael does some explaining about how the FDA and clinical trials work.]

Readers of this blog have a pretty wide range of opinions about the FDA, although the more outspoken of you mostly chime in with negative comments. Doc Gumshoe, being aware of the many stark conflicts between the FDA and drug companies, is a skeptic regarding the view that the FDA is in cahoots with Big Pharma. But this little piece looks at a few areas where the FDA’s recent practices have been subject to criticism from elevated sources. Before I go on to cite chapter and verse, I should recap some of the specific hurdles that the drug and device approval process has to leap over.

What does a treatment have to demonstrate to gain FDA approval?

The answer, in two words (plus a conjunction) is efficacy and safety. That goes for any intervention, whether it’s a drug or a device or any other kind of treatment. But how can that be demonstrated? If we’re talking about a drug, the gold standard is the randomized, placebo-controlled clinical trial, in which the candidate drug demonstrates at least some advantage over an existing drug. The advantage could be with regard to any of several parameters – a greater percentage of patients attaining a specific marker, a more rapid onset of action or a longer duration of action, a lower incidence of adverse effects, and a number of others. And this advantage needs to be statistically significant, meaning that the results of the clinical trial must be highly unlikely to have been a matter of chance.

There are lots of obstacles that the pharma company has to overcome. Here are just a few:

If there are already some fairly effective existing drugs for a disease, it’s likely to be difficult to enroll patients in a trial where the patient would not know whether he/she was going to get the existing drug, or a placebo, or a new experimental drug.

If the trial calls for patients/subjects who have never taken a drug for that disease or condition – drug-naïve patients, as they are known – it may be exceedingly difficult to recruit enough patients for the kind of big Phase 3 trials the FDA likes to see. Patients newly diagnosed with a disease or disorder want to be treated with a drug that is known to work.

Then there’s the issue of diseases/disorders/conditions that don’t affect large numbers of people. Again, in such cases, it may be difficult or impossible for drug companies to enroll enough patients for multiple clinical trials.

For those reasons (and others) lots of trials are now being conducted in the so-called “less developed” parts of the world, and sometimes, perhaps, with less rigorous standards than would be expected.

… and, of course, there’s another issue …

Namely, how long it takes for some drugs to get approved and available for treatment. We all remember the early days of the AIDS epidemic. When there was something like a promising drug coming along, there was intense pressure to get that drug to people who were dying of AIDS. That got a lot of publicity, and the FDA got it in the neck for being “obstructionist” – that is, for insisting on going through the time-consuming approval process.

The FDA, to some degree, initiated changes in their procedures – mostly a matter of streamlining the bureaucracy and simplifying procedures.

But in response to the clear need to speed things along – get drugs to the patients who need them – the FDA has also instituted some changes in the requirements for approval, specifically for some medical conditions, based on such issues as limited availability of treatment options and urgency of getting some kind of treatment to patients who might have rapidly deteriorating illnesses.

“Breakthrough Therapy Designation,” or BTD

Here’s a statement directly from the FDA:

Fact Sheet: Breakthrough Therapies

On July 9, 2012 the Food and Drug Administration Safety and Innovation Act (FDASIA) was signed. FDASIA Section 902 provides for a new designation – Breakthrough Therapy Designation. A breakthrough therapy is a drug:

• intended alone or in combination with one or more other drugs to treat a serious or life threatening disease or condition and
• preliminary clinical evidence indicates that the drug may demonstrate substantial improvement over existing therapies on one or more clinically significant endpoints, such as substantial treatment effects observed early in clinical development.

If a drug is designated as breakthrough therapy, FDA will expedite the development and review of such drug. All requests for breakthrough therapy designation will be reviewed within 60 days of receipt, and FDA will either grant or deny the request.

Please refer to section 902 of FDASIA for more specific information about this provision. Additionally, a draft Guidance for Industry: Expedited Programs for Serious Conditions––Drugs and Biologics was published on June 25, 2013, and includes information such as qualifying criteria for breakthrough therapy designation, features of the breakthrough therapy program, and guidelines on how to submit a breakthrough therapy designation request.

As of the beginning of 2014, the FDA had received over 100 breakthrough therapy designation (BTD) requests and approved only about 30%. They declined about 60% and another 10% or so are currently pending.

Some of the “winners” are:

• Gazyva (obinutuzumab), developed by Genentech (Roche), for chronic lymphocytic leukemia (CLL) in combination with chlorambucil. CLL is not a rare disease. Indeed it’s the most common form of leukemia, and in some patients it progresses so slowly that it may be reasonable not to treat at all. However, there are more rapidly-progressing variants that warrant more aggressive treatment, and Gazyva’s promise is that it will be less toxic than some of the current combined therapies.
• Imbruvica (ibrutinib), developed by Janssen and Pharmacyclics (Johnson & Johnson), for mantle cell lymphoma, a genuinely rare form of cancer for which effective treatments options have been lacking.
• Sovaldi (sofosbuvir), from Gilead, for hepatitis C. The advantage here is that it allows for treatment without interferon. Treatment with interferon may need to continue for a year, with nasty side effects, and the cure rate with Sovaldi approaches 100%.

So those are a few successes for BTD, approved in near-record time. However, this supposed pathway to earlier drug approval has not by any means been the bonanza the pharmaceutical industry was hoping for. The majority of the new drug applications still need to go the slow way, and the slow way is still slow.

But the FDA is also getting criticized for being too lax!

A JAMA paper (Downing NS et al, “Clinical trial evidence supporting FDA approval of novel therapeutic agents 2005 – 2012,” JAMA 2014;311(4):368-377), published a few weeks ago, examined the basis on which new drug treatments were approved by the FDA between 2005 and 2012. A total of 188 new drugs were approved, for 206 indications. These approvals were based on 448 so-called “pivotal” trials, meaning relatively large Phase 3 trials. However, the investigators reported several findings of concern:

• Of the 188 new drug approvals, 74 were based on the results of a single pivotal trial.
• 91 – almost half – were based on surrogate end points rather than on clinical evidence of improvement in the patients’ disease or disorder. And an additional 36 were based on scores on tests used to measure the impact of a disorder, such as depression or cognition. Only 67 were based on clinical outcomes.
• Only about 40% of the trials compared the new agent with an existing drug; the rest compared the new agent only with placebo.

Granting approval to a drug based on a single trial might be justified based on the importance of making the drug available to patients for whom other options are significantly less effective, or if there are impediments in mounting another large Phase 3 trial – i.e., difficulty in recruiting patients.. However, a fundamental scientific principle is that results of experiments – and a clinical trial is definitely an experiment – have got to be able to be replicated. The history of science is rife with examples of experiments that could not be replicated: some of us may remember the announcement that scientists had been able to achieve “cold nuclear fusion” – a pipe dream that has never been replicated. A single Phase 3 trial is all very well, but don’t we want more evidence than that?

A “surrogate end point” is an easily-assessed measurement of some parameter that stands in for the real clinical goal of the treatment being evaluated. Employing surrogate end points might be justified on the basis of pure practicality. For example, a frequent surrogate end point in trials of cancer drugs is measuring tumor size. If the tumor shrinks, the drug is assumed to be working. There’s an obvious problem with waiting for the more definitive end point, that being the disappearance of the tumor, or alternatively, the death of the patient. But tumors have been known to shrink and grow again, so it may be somewhat premature to approve a drug for widespread use based on that particular surrogate end point.

Similarly, lowering blood pressure and lowering cholesterol are surrogate end points for lowering the risk of heart disease, and by and large the evidence is convincing that in patients who are at significant heart disease risk, lowering those parameters does seem to translate into decreased heart disease mortality. However, we need to remember that for a very long time after cholesterol had been fingered as the culprit in atherosclerosis, all of the experiments to determine whether lowering cholesterol actually conferred a survival benefit were inconclusive – yes, the interventions lowered cholesterol, but it made no difference in patient survival. It was only in 1994, with the Scandinavian Simvastatin Survival Study, that a clear and definitive survival benefit was seen. But that survival benefit, quite clear in “at risk” patients, is not so clear in the general population.

Finally, failing to conduct clinical trials where the new agent is compared head-to-head with an existing agent, and instead comparing the new agent only with placebo, is definitely the preference of the pharmaceutical companies, but not likely to be in the public interest. One could see how the pharma outfits would prefer trials where their new drug is compared with placebo rather than with an existing drug which is known to be safe and effective. In a trial versus an existing drug, they might come in second, and that would be curtains, while in a trial versus placebo, they have at least a chance to try to spin the results in the most positive light. But from the patients’ perspective, the comparison with the existing agent is clearly the more valuable information.

The JAMA paper is not about the breakthrough drug designation, which only got underway at the very end of the period covered in the paper. But it points at the underlying tension between two (at least two!) high priorities – one, to make sure that drugs approved by the FDA are indeed safe and effective, and two, to make sure that patients who can benefit from these drugs don’t languish in their illness for an unnecessarily long time.

By the way, breakthrough drug designation is not the same as orphan drug designation, which gives pharmaceutical companies incentives to develop drugs for diseases which affect fewer than 200,000 persons – so-called “orphan diseases.” The objective of BTD is getting drugs to patients quickly, of the orphan drug designation, developing drugs for treatment areas where the pharmaceutical companies may not have big financial incentives.

So, how do we arrive at a procedure that addresses both issues?

What The Celebrated Man (or Woman) In The Street (T. C. MITS / WITS, as he/she were referred to in some excellent books I read in my early youth) wants and needs is reliable assurance from High Places that the drug he/she needs is definitely okay. But at the same time MITS and WITS want the drug when they need it, and not five years later. So what’s to be done?

Doc Gumshoe’s answer is, make the drug available, but tell MITS/WITS “here’s what we know about it so far,” with emphasis is on “so far.” The drug developer and the FDA have got to keep gathering data (which they’re absolutely supposed to do – it’s called “post-marketing surveillance”), and the FDA needs to keep reviewing the status of any drug that was granted approval based on less than the usual standard. Yes, if the results of that single Phase 3 trial versus placebo warrant it, approve the drug, but require further data, and keep a watchful eye on that drug.

T. C. MITS/WITS should understand that FDA drug approval does not constitute the untarnishable Gold Star that forever guarantees that the drug is safe and effective. It just means that based on the available evidence to date, the FDA considers that it’s okay to prescribe the drug to patients who need it. For example, the FDA would not approve some drugs, such as cancer drugs that have toxicities, for the treatment of hay fever. (That doesn’t mean that physicians might not prescribe it for hay fever, or for any other condition – that’s called “off-label prescribing,” and physicians are permitted to do it. However, if a patient has a bad adverse effect due to off-label prescribing, the pharmaceutical company is off the hook, and the physician bears the entire responsibility.)

But the FDA needs to be more assiduous about demanding and evaluating data after drug approval, and needs to be proactive about getting further information to the public – healthcare providers and patients alike.

And, on another issue, FDA approval of medical devices …

Evaluating the efficacy and safety of medical devices might appear to be a good deal simpler. After all, it either works or doesn’t work, and it either harms the patient or it doesn’t, right? But there are inherent complications. Device manufacturers adopt complex (and expensive) stratagems to come up with “dummy” devices that look like the real devices, but don’t do anything, so as to mimic the placebo arm of a drug trial. That works sometimes. But if the device is something like a heart valve, or an artificial hip, dummy devices are out of the question.

So lots of devices are approved on the basis of observed probable effectiveness, and this is particularly the case when the patient population for whom this device is intended is fairly small. A case in point is a device, the BSD-2000 (made by BSD Medical in Salt Lake City), that employs microwaves to raise the body temperature in cancer patients, which supposedly makes cancer cells more susceptible to radiation therapy. It was originally developed to treat patients with advanced cervical cancers that did not respond to other treatments, but has since been used “off-label” to treat other cancers.

Devices of this type can be quite expensive – the BSD-2000 runs half a million dollars – and this obviously creates a considerable incentive for facilities that own such devices to use them, perhaps off label. And because there is an FDA provision called “humanitarian device exemption” that applies to devices for disorders affecting fewer than 4000 patients, allowing for approval to be granted based on a presumption of effectiveness, which opens the door for more widespread use of devices, perhaps for conditions other than those for which they were originally approved.

Does relaxing the approval process in this way provide a benefit to some patients? Clearly, yes! But does it perhaps also open the door to abuse? Perhaps so. We know that when medical facilities invest in some imaging devices, it’s hard to get out the door without being scanned. Pain in your pinky? Into the MRI!

And a warning about “FDA Warnings”

A valued reader sent a link to a site with the scary headline “20 Drugs You Should Not Be On!” This was based on FDA warnings in the prescribing information (PI) for those drugs. Here’s the list, with a brief summary of the warnings:

A 2007 federal law requires the FDA to disclose all its investigations into reports of possibly drug-related adverse events. The first of this series of quarterly reports has been released. It includes adverse events reported between January 1 and March 31, 2008.

Here’s the list of drugs and the adverse events that have been reported:

Arginine Hydrochloride Injection (R-Gene 10) — Pediatric overdose due to labeling and packaging confusion
Desflurane (Suprane) — Cardiac arrest
Duloxetine (Cymbalta) — Urinary retention
Etravirine (Intelence) — Hemarthrosis (blood in a joint)
Fluorouracil Cream (Carac) and Ketoconazole Cream (Kuric) — Adverse events due to name confusion
Heparin — Anaphylactic-type (life-threatening allergic) reactions
Icodextrin (Extraneal) — Hypoglycemia (low blood sugar)
Insulin U-500 (Humulin R) — Dosing confusion
Ivermectin (Stromectol) and Warfarin — Drug interaction
Lapatinib (Tykerb) — Hepatotoxicity (liver toxicity)
Lenalidomide (Revlimid) — Stevens-Johnson syndrome (a deadly drug reaction)
Natalizumab (Tysabri) — Skin melanomas (deadly skin cancer)
Nitroglycerin (Nitrostat) — Overdose due to labeling confusion
Octreotide Acetate Depot (Sandostatin LAR) — Ileus (bowels not moving)
Oxycodone Hydrochloride Controlled-Release (OxyContin) — Drug misuse, abuse, and overdose
Perflutren Lipid Microsphere (Definity) — Cardiopulmonary reactions (lung/heart problems)
Phenytoin Injection (Dilantin) — Purple glove syndrome (discoloration, pain, and swelling of the hand that may lead to amputation)
Quetiapine (Seroquel) — Overdose due to sample pack labeling confusion
Tebivudine (Tyzeka) — Peripheral neuropathy (tingling or numbness in the extremities)
Tumor Necrosis Factor (TNF) Blockers — Cancers in children and young adults

Are those warnings real? Yes indeed. But do they mean that “you should not be on those drugs?” For most patients, the warnings don’t mean anything more than “be careful,” which one should always be when taking drugs of any kind. To rule out nitroglycerin due to “labelling confusion” would deprive huge numbers of people with angina from very safe and highly effective relief – and likely consign them to more invasive treatment. To rule out fluorouracil and ketoconazole because of confusion between Carac and Kuric is dopey! Should diabetics be deprived of Humulin because of the possibility of dosing confusion? Well, I won’t go on.

The scoop on FDA warnings is that there’s hardly a single drug in the entire Physician’s Desk Reference, which contains all the PIs for all FDA approved drugs, that doesn’t include a warning. There are always warnings. That doesn’t mean we don’t have to take them seriously, but it also doesn’t mean that we have to run screaming in the opposite direction.

* * * * * * *

Thank you for all the comments; keep them coming! By now, hardly anyone hasn’t heard about the big retrospective study that concluded that mammograms don’t result in any survival benefit. I’ll have quite a lot more to say about that, but for now, I will just take a few shots:

• One, it’s a 25 year retrospective study, employing radiographic techniques back to the mid-1980s.
• Two, it excluded ductal carcinoma in situ (DCIS), on the grounds that those cancers “rarely need treatment.” DCIS cannot be detected by manual breast examination, and a significant percentage of these cancers do eventually become invasive.
• Three, a conclusion was that mammograms resulted in overdiagnosis, and “unnecessary treatment” of about one in three diagnosed cancers.

But how does anyone know whether it’s necessary to treat the cancer, unless and until the patient dies of some other cause?

Best to all, Michael Jorrin (aka Doc Gumshoe)