Benton Bramwell, ND, and Matt Warnock, JD

Historical Context

In response to a real need to enhance FDA’s ability to assess new drug applications, The Prescription Drug User Fee Act, allowing FDA to collect fees from industry in order to provide resources needed to review drug applications, was passed by Congress in 1992.

In 1994, in issuing its guidance to industry about which study designs should be utilized to gather dose-response data moving forward, the International Conference on Harmonisation of Technical Requirements for the Registration of Pharmaceuticals for Human Use (ICH) assessed the inherent difficulties of trying to adequately derive dose-response data by stating, in part:

Historically, drugs have often been initially marketed at what were later recognized as excessive doses (i.e., doses well onto the plateau of the dose-response curve for the desired effect), sometimes with adverse consequences (e.g., hypokalemia and other metabolic disturbances with thiazide-type diuretics in hypertension). This situation has been improved by attempts to find the smallest dose with a discernible useful effect or a maximum dose beyond which no further beneficial effect is seen, but practical study designs do not exist to allow for precise determination of these doses. Further, expanding knowledge indicates that the concepts of minimum effective dose and maximum useful dose do not adequately account for individual differences and do not allow a comparison, at various doses, of both beneficial and undesirable effects. Any given dose provides a mixture of desirable and undesirable effects, with no single dose necessarily optimal for all patients.

Despite recognition of difficulties in adequately describing dose response and guidance to help inform dose response research from ICH, and despite a greater magnitude of resources available to FDA to better evaluate drug applications, in 2014 it was estimated that about a quarter of drugs receive a black box warning or undergo market withdrawal after approval, which is an increase from the rate prior to 1992 (26.7 per 100 drugs post 1992 vs 21.2 per 100 drugs before 1992). While there is more than one simple reason for the increase in rate of post-market safety problems with approved drugs, we feel moved to ask: if even with helpful guidance to improve dose-response data gathering and greater regulatory resources to evaluate drug safety the rate of safety concerns remains high (and even increases), do we not have a fundamental problem with the unsafe use of drugs entering the marketplace?

Closing the Information Gap in Dosing

One of the roots of this problem is surely the often-significant gap between the dose of a drug that produces a desired response in half the population, ED50, and the actual doses of drugs approved and then prescribed in the marketplace, an issue partly addressed in our first article. A further example of the dilemma caused by the delta between ED50 and doses used in practice is the case of aspirin used to prevent myocardial infarction. While doses as high as 1300 mg/d have been approved in the US to prevent cardiovascular disease, doses as low as 30 mg/d are shown to fully inhibit platelet thromboxane production. While a more reasonable dose of 81 mg/d is in the recent past most commonly used, a dose of 325 mg/d still has represented the second-most commonly used dose in 35% of cases. This is unfortunate as data from observational studies did not support any increase in efficacy beyond 75-81 mg/d, while larger doses were shown to increase the risk of bleeding.

Granted, when aspirin first appeared in market its use was not as a drug to lower cardiovascular risk (its use to offset risk of myocardial infarction was cemented in the 1989 Physician’s Health Study), but given that off-label use of drugs is still a common practice, the example is still very fitting. If we only learn what actually constitutes a usually safe and effective dose for aspirin many years into its use, how far are we from truly understanding the safety of the rest of the modern pharmacopeia?

While we and everyone else probably has a whole lot to learn about the safety profile of drugs now on the market, one positive step might simply be to require drug marketing advertisements and prescribing information provided to patients through pharmacies to disclose any available ED50 for the use at hand, along with the context of what ED50 means, as well as any doses approved for the use prescribed. How different would be the conversation between patient and provider be if truly informed patients asked the question: “Why do I need to take a dose so much above the ED50?” And, if prescribers don’t actually know all the safety nuances of use at doses well into the plateau of the dose-response curve, why—except in cases where only maximal benefit would be enough to save a patient’s life—would one want to start with aggressive dosing?

Another problem leading to fuzziness in clinical decision making while trying to take into account both drug efficacy and toxicity is the reality articulated in the ICH statement above that individual differences exist in drug efficacy and drug toxicity. While dose response data better brings into focus the central tendency of data gathered from a population, differences in genes encoding for drug metabolizing enzymes, variations in body size and composition, diet, microbiome composition, and other concurrent medication use are among the factors that may affect the pharmacokinetics and pharmacodynamics of a drug. The statistics generally used to describe dose responses are usually applied to data from a population, not an individual. However, it is the unique response of any given patient that really matters at the moment they are the focus of clinical care: odds of what adverse events may or may not occur in a population are eclipsed by any adverse outcome that turns out to occur with 100% certainty in a particular patient.

Moreover, the lack of homogeneity caused by individual variability also exists with another factor complicating the interpretation of safety data: the change in the signal to noise ratio as one moves from lower to higher doses. That is, less immediately life-threatening toxicities, especially those that may manifest when a drug is started and then wane in the short term, are more difficult to identify at lower doses than severe toxic effects that occur more quickly at larger doses. With relatively lower signal and higher noise at lower doses, it is difficult to determine when toxicity first really begins to declare itself. Also, while it is advantageous financially to structure a study with as few subjects as necessary in order to see an effect tied to efficacy, there is no guarantee that a study with enough power to detect a desired efficacious effect will also have enough subjects to detect the most important safety signals. Larger studies reduce the chance of statistical error of missing a relevant toxicity effect. A more balanced approach in study design would be to power studies to detect both efficacy in an outcome of efficacy and at least what can be anticipated as the most likely/concerning adverse event.

Given the mix of both desirable and undesirable effects at any given dose studied in a population, and the changes in signal to noise ratio that exist along any dose response curve, perhaps those trying to exactly predict an optimal dose for an individual patient may wish to temporarily change places with physicists trying to pinpoint the exact location of any given particle and then frankly discuss which is the more fruitful of the two exercises.

The Variable of Time

While what we have written above clarifies some of the very real challenges confronting the usage of drugs in the United States and the world, it should not be taken as a blanket indictment of motive. Aspirin, a drug on the market for many years, was found to have an additional action of reducing risk of myocardial infarction. That’s a good thing. But if in our need to quickly address a pressing health concern we intervene without knowing as much as possible about trade-offs in safety, then we really need to have a strong system to flesh-out the more complete safety picture over time.

Currently, FDA can and does require/catalyze a number of post-marketing studies through postmarketing requirements (PMR, a study a drug company is required to do) and postmarketing commitments (PMC, a study a drug company commits to do) when a drug or biologic is approved. In fact, about 80% of newly approved therapeutics (both drugs and biologics) are required to perform at least 1 clinical study, which about 60% of the time takes the form of a prospective cohort study, registry, or clinical trial. This is good progress in terms of the number of postmarketing studies being done, but given that most drugs are approved based on several studies of up to only a few years each, and that some adverse effects may only manifest after 5+ years of use, we reiterate the observation of others that a critical factor of postmarketing studies is the length of time that these studies are performed.

An example of just how critical the variable of time is to determine drug safety is illustrated by the work of the DAD study group, which in a prospective observational study of over 20,000 patients with HIV found that the incidence of myocardial infarction for those taking protease inhibitors over 6 years was 6.01 per 1,000 person years while for those patients not exposed to protease inhibitors the incidence was 1.53 per 1,000 person years. This is incredibly important information that would have been missed if such a long-term prospective study had not been done.

It seems that in terms of really understanding drug safety, quantity of time studied has a quality all its own. Due to the pragmatic challenges of prospective clinical trials done over this length of time, such as patients wanting to try different medications or not wanting to possibly take placebo for many years, it makes sense to uniformly require that after drugs obtain approval, they be monitored in a prospective observation study for about seven years so that the risk of adverse effects over the long-term can be more fully identified. This is an active approach to safety data gathering that would greatly increase the safety data collected in conjunction with the passive reporting currently collected as reports are submitted through the MedWatch system.

Of course, we cannot help but point out that with a more complete picture of longer-term safety of newly approved, single drugs being elusive, the relative safety of low-dose combination therapy seen in shorter term studies that we have previously highlighted suggests that this approach will in many cases come to be clearly understood as the superior approach for long-term safety as well.

References

1. . https://www.fda.gov/industry/prescription-drug-user-fee-amendments/pdufa-legislation-and-background Accessed 19 October 2022.
2. . Guideline for Industry: Dose-response information to support drug registration. International Conference on Harmonisation of Technical Requirements for the Registration of Pharmaceuticals for Human Use. 1994. Accessed 19 October 2022. https://www.fda.gov/media/71279/download
3. . Frank C, Himmelstein DU, Woolhandler S, Bor DH, Wolfe SM, Heymann O, Zallman L, Lasser KE. Era of faster FDA drug approval has also seen increased black-box warnings and market withdrawals. Health Aff (Millwood). 2014;33(8):1453-9.
4. . Campbell CL, Smyth S, Montalescot G, Steinhubl SR. Aspirin dose for the prevention of cardiovascular disease: a systematic review. JAMA. 2007 May 9;297(18):2018-24. doi: 10.1001/jama.297.18.2018. PMID: 17488967.
5. . Steering Committee of the Physicians’ Health Study Research Group. Final report on the aspirin component of the ongoing Physicians’ Health Study. N Engl J Med. 1989 Jul 20;321(3):129-35. doi: 10.1056/NEJM198907203210301. PMID: 2664509.
6. . Skydel JJ, Zhang AD, Dhruva SS, Ross JS, Wallach JD. US Food and Drug Administration utilization of postmarketing requirements and postmarketing commitments, 2009-2018. Clin Trials. 2021;18(4):488-499. doi:10.1177/17407745211005044
7. . Resnik DB. Beyond post-marketing research and MedWatch: Long-term studies of drug risks. Drug Des Devel Ther. 2007;1:1-5. doi:10.2147/dddt.s2352
8. . DAD Study Group, Friis-Møller N, Reiss P, Sabin CA, Weber R, Monforte Ad, El-Sadr W, Thiébaut R, De Wit S, Kirk O, Fontas E, Law MG, Phillips A, Lundgren JD. Class of antiretroviral drugs and the risk of myocardial infarction. N Engl J Med. 2007 Apr 26;356(17):1723-35. doi: 10.1056/NEJMoa062744. PMID: 17460226.

Published August 12, 2023

About the Authors

Benton Bramwell, ND, is a 2002 graduate of National College of Naturopathic Medicine who practiced primarily in Utah while helping to expand the prescriptive rights of naturopathic physicians in that state. Currently, he owns and operates Bramwell Partners, LLC, providing scientific and regulatory consulting services to both dietary supplement and conventional food companies. He and his wife, Nanette, have six children and two grandchildren; they live in Manti, Utah.

Matt Warnock is an accidental herbalist, who received his MBA and Juris Doctor from BYU, then worked as an attorney, litigator, and business consultant until 2000. He then joined RidgeCrest Herbals, a family business started by his father, and started learning about herbal medicine, focusing especially on complex herbal formulas. He has two U.S. patents for herbal formulations and methods. He lives near Salt Lake City with his wife, Carol; they are the parents of three children and four grandchildren.