Unpacking the Data from Merck & Co. Clinical Trials
October 16, 2018
The following information and conclusions are based on opinions formed after a review of relevant facts and data by John Ray, Senior Consultant, Mass Tort Nexus
Herpes zoster (shingles) is a symptom of the varicella-zoster virus, the same virus that causes chickenpox. Individuals experiencing active herpes zoster infections often report debilitating pain and blistering skin rashes typically located on the face and torso. The varicella-zoster virus (chickenpox) can remain dormant in the body indefinitely and may emerge decades later as herpes zoster (shingles). Not all persons who have the varicella-zoster virus will develop herpes zoster, and it is impossible to predict who will and who won’t.
American pharmaceutical giant, Merck & Co. (Merck), released its Zostavax vaccine to the market in 2006. In its marketing campaign, Merck claims that “you have a 1 in 3 chance of contracting shingles,” and that the Zostavax vaccine reduces the occurrence of zoster by “51% overall in subjects aged 60 years or older.” The FDA approved the Zostavax vaccine prior to its market release.
Summary of Findings
- At a minimum, valid clinical trial results are based on two conditions: (1) trial participants must be diagnostically homogeneous (i.e., they share the same medical diagnosis); and, (2) the number of trial participants must be statistically significant (i.e., results may be reliably extrapolated to a larger group).
- Results from the Zostavax clinical trials conducted by Merck are arguably invalid because the trial participants were neither diagnostically homogeneous or comprised a statistically significant number.
- Because the risk of developing herpes zoster is higher for those that received the Zostavax vaccine than those who didn’t, Merck’s public marketing campaign of the Zostavax vaccine misled consumers, at best, and caused significant harm, at worst.
Methodology Flaws in the Zostavax Clinical Trials
Diagnostically Homogeneous Trial Participants
Generally, clinical trials are intended to observe occurrences and outcomes from product administration with trial participants that share a medical diagnosis (i.e., diagnostically homogenous). For example, a clinical trial involving a drug to treat type 2 diabetes would only enroll individuals diagnosed with type 2 diabetes. Once a diagnostically homogenous group is identified and recruited for a clinical trial, an equal proportion of the group will be administered the drug being studied while the remainder will be administered either a placebo or another drug intended to treat the same disease or condition in trials testing one drug against another.
Zostavax vaccine clinical trials conducted by Merck differed significantly from normal clinical trial methodology in that Zostavax is not intended to treat an existing diagnosed disease or condition but rather is intended to prevent herpes zoster, a symptom of the varicella-zoster virus. If trial participants did not share a diagnosis of an existing disease or condition and could not reliably predict the future onset of a herpes zoster condition, trial participants were not—and could not be—diagnostically homogeneous. Therefore, conclusions reached by Merck from the Zostavax clinical trials are arguably invalid.
Statistically Significant Number of Trial Participants
After clinical trial observations are recorded, the results are extrapolated to represent predictable outcomes in a larger population. Extrapolations from statistically significant numbers are never as reliable as data collected from a larger, real-world population. Significant real-world data (e.g., data collected from Medicare, countries with national health care systems and/or unbiased third-party authorities) existed prior to the Zostavax trials regarding occurrence rates of herpes zoster by age group.
Numerous authorities and governments had already established that the probability of contracting herpes zoster during a lifetime was between 30 and 33%. Given the fact that a “risk without vaccination rate” was generated using far more reliable methods, there was no reason for Merck to include a placebo group in their trials when real-world data existed relevant to the unvaccinated.
In addition, Merck disqualified participants with compromised immune function from the Zostavax clinical trials. Immuno-compromised individuals are significantly more likely to develop herpes zoster. The 30 – 33% lifetime occurrence rate established by trusted sources from real-world data included persons with compromised immune function. As a result, the real-world data must be considered more accurate than the Zostavax trial data relevant to the placebo group.
The following chart shows the broad results of the first Zostavax trial conducted by Merck. Merck claims that for subjects 60 years of age or older there is a 51% reduction of risk. These results are misleading because they average occurrence rates of persons 80 years of age and older (who are significantly more likely to develop herpes zoster based on real-world data) with occurrence rates from age groups that are significantly less likely to develop herpes zoster. These statistics were used by Merck to obtain licensure for Zostavax from the FDA.
[Merck’s Zostavax link: https://www.merckvaccines.com/Products/Zostavax/efficacy/]
Merck’s Misleading Marketing Scheme
Merck used the blanket statement, “You have a 1 in 3 chance of contracting shingles,” in their fear-based advertising campaigns for Zostavax. If we ignore the placebo trial group data in favor of the more reliable real-world data which tells us that 33% of individuals will experience herpes zoster during their lifetime leaving 67% that will not, we can conclude the following:
Age Group 60-69
Merck claimed a Zostavax efficacy rate of 64%. When this rate is compared to the real-world data rate, we can conclude that administration of the Zostavax vaccine increased the risk of experiencing herpes zoster by 3% for this group. Those who do not receive the Zostavax vaccine have a 67% chance of never manifesting herpes zoster symptoms while those that do receive the vaccine have a 64% chance of never contracting Herpes Zoster.
Age Group 70-79
Merck claimed a Zostavax efficacy rate of 41%. When this rate is compared to the real-world data rate, we can conclude that administration of the Zostavax vaccine increased the risk of developing herpes zoster by 26% for this group. Those who do not receive Zostavax have a 67% chance of never manifesting herpes zoster symptoms while those that receive the vaccine have a 41% chance of never contracting Herpes Zoster.
Age Group > 80
Merck claimed a Zostavax efficacy rate of 18%. When this rate is compared to the real-world data rate, we can conclude that administration of Zostavax increased the risk of developing herpes zoster by 49% for this group. Those who do not receive Zostavax have a 67% chance of never manifesting herpes zoster symptoms while those that receive the vaccine have an 18% chance of never contracting Herpes Zoster.
The blended rate (all age groups in the study combined) of 51% efficacy claimed by Merck compared with occurrence rates from real-world data, leads us to conclude that for all intended users, the risk of developing herpes zoster after vaccination with Zostavax is greater than prior to vaccination. Real-world data demonstrates that the relative risk of contracting herpes zoster post-vaccination is 49% while those who are not vaccinated face a 33% risk.
The following graph shows herpes zoster occurrence rates by age group. A comparison of Zostavax trial data to real-world occurrence rates supports another conclusion—the age groups most at risk for developing herpes zoster (and most in need of an effective vaccine) had the least probability of protection from administration of the Zostavax vaccine and were arguably at the highest risk for developing Zostavax as a result of receiving the vaccine.
The foregoing is an observation of statistics and data related to Zostavax. The method by which Merck used and manipulated this data in misleading marketing and advertising is covered in other sections of the material.
The conclusions contained herein are based on opinions formed by the author after a review of the relevant data. We acknowledge that others could draw differing conclusions and opinions based on the same observations.