FAQs

Administration of thimerosal-containing vaccines to infant rhesus macaques does not result in autism-like behavior or neuropathology. S. Bharathi et al.
PNAS, 2015 ; published ahead of print September 28, 2015
http://www.pnas.org/content/early/2015/09/24/1500968112.abstract

FREQUENTLY ASKED QUESTIONS

Updated November, 2015


Why was this study done?
In light of concerns expressed by some members of the community over the safety of pediatric vaccines, researchers examined the safety of a variety of vaccine schedules, including the current pediatric vaccine schedule, which has greatly expanded over the last decade.


What was this study investigating?
This study was designed to examine the safety of US-recommended pediatric vaccine schedules in non-human primates (rhesus macaques). Different groups of animals received vaccines like those used in the 1990s; the measles-mumps-rubella (MMR) vaccine; or vaccines like those used in the current recommended US vaccine schedule. Several of the pediatric vaccines manufactured in the 1990s contained thimerosal, whereas most current vaccines do not. (The MMR vaccine has never contained thimerosal.) Vaccinated and unvaccinated macaques underwent behavioral testing from 12 to 18 months of age. At the end of the study, researchers examined the cellular structure of several brain regions that are altered in autism.1


What is thimerosal?
Thimerosal is an ethyl mercury-containing organic compound. It is used as a preservative in a number of biological products to help prevent contamination from harmful microbes. It has been used in many multi-dose pediatric vaccines, including Hepatitis B, Diphtheria-Tetanus-acellular Pertussis (DTaP), and Haemophilus influenzae B (Hib), as well as some flu vaccines. Thimerosal has now been removed from most vaccines recommended for pediatric use in the US, though it is still used as a preservative in all multi-dose flu and meningitis vaccines.


What were the findings of this study?
Perhaps the most significant finding, with respect to the hypothesis that thimerosal-containing vaccines may have an adverse effect on behavior, was that all animals in the study developed the expected repertoire of social behaviors typically found in rhesus macaques of this age. Virtually no negative behaviors, such as rocking, self-clasping, and stereotypy (repetitive behaviors), were reported, regardless of vaccination status. In fact, the data from the negative behavior categories had to be summed and reported together since there were so few occurrences. Analysis of cell number, size, and density of the specific brain regions studied revealed no differences between vaccinated and unvaccinated animals.


Does this answer the question of whether vaccines cause autism?
This non-human primate study of vaccine safety was not designed to determine whether vaccines cause autism. It should also be noted that in all hypothesis testing, a failure to detect an effect doesn’t mean the effect isn’t there - there simply isn’t supporting evidence in the extant data. In this study there was no evidence of aberrant social behavior or neuropathology in rhesus macaques following vaccination with thimerosal-containing vaccines (TCVs). In light of the concern that TCVs may be associated with autism, the results from this paper are encouraging since social development and the neuroanatomy of the brain regions studies were normal. All of the animals included in this study had a normal birth weight and gestational age. Low birth weight and decreased gestational age have been associated with increased autism risk. Furthermore, there have been a number of studies that suggest that environmental factors during pregnancy may contribute to autism risk. These and other factors were not addressed by this study.


What brain regions were studied?
Analyses were performed in three brain regions known to exhibit neuropathology in postmortem samples from people with autism. In autism, the cerebellum and the lateral nucleus cells of the amygdala have been associated with a reduction in cell number, whereas the CA1 region of the hippocampus has been associated with a reduction in cell size. A number of additional brain regions have been implicated in the etiology of autism. Many additional primate tissues have therefore been collected and banked for future analyses.


What is stereological analysis and why was this method used?
Stereology is the three-dimensional interpretation of two-dimensional cross sections of tissues. In this study researchers used well-accepted unbiased stereological procedures to estimate neuronal number, size, and density of several brain regions. The total number of neurons was quantified using the optical fractionator method, which avoids the well-known problem associated with using neuronal density to estimate total numbers of neurons. The Cavalieri method was also used; this is a well-established stereological technique that uses interpolation between samples to estimate volume of the specific structures being studied. In the cerebellum, for example, every 10th section, 0.6mm apart was examined in the entire structure. The total volume between vaccinated and unvaccinated animals in the whole cerebellum was then compared, as well as by left and right hemispheres. If there had been a hypoplasia of cerebellar vermal lobules VI and VII (as has been identified in autism2) this would have been evident, particularly since these lobules make up the entire superior posterior lobe and almost 50% of the entire primate cerebellum. All of the stereology data were analyzed by at least two researchers at two different academic institutions. Further details on the stereological analyses are available in the Supplementary Information published alongside the paper. A longitudinal MRI analysis of brain structure is currently underway. This will enable the researchers to determine whether the volume of specific structures changes over time as animals age and receive additional TCVs.


What is a blinded study? Other than the researchers involved, who else played a role in the analysis or interpretation of data in this paper?
All of the assessments were conducted “blinded,” meaning that the technicians, behaviorists, and researchers involved in the collection of data did not know which study group an animal was assigned to. Once the data were collected a “chain of custody” protocol was implemented to provide chronological documentation showing the control, transfer, and analysis of all datasets, and an independent statistical consultant was used for all analyses. The statistical analyses were first performed on a “group” basis, meaning the statistician analyzed 6 groups of data labeled 1-6, without knowing the group assignment in the study. As scientific researchers with many decades of experience between them, the researchers believe that everything was done to ensure the integrity of the data. We can appreciate that some of the organizations that contributed funding to this study may be disappointed that the findings were not what they hoped or anticipated. The study authors have a responsibility to report the results thoroughly and accurately - that is what they did, without interference.


Why were the neuroanatomical studies only conducted in 3 groups of animals while the behavioral studies were conducted in all groups?
The neuroanatomical analyses were first performed in brain sections from the 1990s Primate and 2008 groups, as animals in these groups received the highest amount of thimerosal exposure (1990s Primate) or the most extensive vaccine exposure (2008). When compared with similar tissue sections from control animals, no neuronal differences were observed so cellular analyses of all vaccine groups were not fully studied.


Why did you assess behavior in animals from 12-18 months of age only?
The researchers have previously reported on early developmental, cognitive, and behavioral measures in this same cohort of animals from birth to 12 months of age.3


Did the paper discuss the fact that the Nonsocial Explore data for the control group was significantly different from some of the vaccinated groups?
Yes, as described in the paper the only instance of a significant effect involving “group” was for Non-Social Explore behavior (the most common primate behavior demonstrated by healthy macaques of this age). Animals in the control group exhibited significantly more Nonsocial Explore behavior at the beginning of social living compared with some of the vaccine groups, but not with the 2008 vaccinated group. This group also received an in utero exposure and multiple infant exposures to a TCV (influenza vaccine), and received the highest number of pediatric vaccines. Therefore, there is no plausible explanation based on vaccination exposure status that correlates with this one significant finding at 12 months of age, and there were no significant differences in any behavior measured between the control and vaccine groups after 6 months of social living (approximately 18 months of age).


If the rates of autism are currently 1 in 68 boys, how can a study with a sample size of only 12-16 animals per group provide enough power to assess behavioral changes?
This study was not designed to determine whether vaccines cause autism and, as such, one cannot rely on the estimated rates of autism in the U.S. (1 in 68 boys) to determine sample size. Indeed it would take several thousand subjects for a study to detect such rare occurrences. The study is best suited to test the hypothesis that a trend across all animals would be observed. Variable transformations, such as the log transformation, are commonly implemented prior to fitting linear statistical models. Linear models are not designed to test for rare events or extreme values.

In order to understand the sample size requirements for this study, it is necessary to first consider the context for interpreting small, medium, and large effect sizes. Small effect sizes are described as “not so large as to make them fairly perceptible to the naked observational eye,” medium effect sizes are described as “large enough to be visible to the naked eye,” and large effect sizes are described as “grossly perceptible.”4 In this study, the goal was to detect obvious behavioral differences (i.e. large effects). Using what is considered a standard effect size measure, the power analysis for this study indicated that 15 animals per group would be needed for comparing means between multiple groups. The study was adequately powered to detect hazard ratios as small as 1.91 with a power of .80, for a 2-tailed α = .05 in proportional hazard models.


Who funded this study?
The vast majority of funding for this study was provided by The Ted Lindsay Foundation and the Johnson family. Smaller grants or donations were provided by SafeMinds, National Autism Association, and the Vernick family. This work was also supported by WaNPRC Core Grant RR0166 and CHDD Core Grant HD02274.


Will non-human primate tissues be available for collaborators to study?
Material Transfer Agreements are currently being developed to enable the various datasets and biological samples derived from this study to be licensed to other collaborators. These will be made available to researchers once publication of study data has been completed.


In a recent blog article, an advocacy organization that contributed funding to this research project indicated that they believed that a part of the study they had agreed to fund had never been conducted. The blog stated that “…the scientists chose not to look at the brains of the primates from this arm of the study.” Is this true and, if so, why wasn’t this part of the study done?
A grant proposal was submitted to this advocacy organization in 2012 requesting $50,000 to section, stain, and perform the stereology of brain sections from an additional 12 animals that had been added to the study in 2011. This new group of animals received the same TCVs as the 1990s Primate group, but on a similar vaccine schedule to children (i.e., not the accelerated schedule used for all of the other vaccine groups). Animals in this group (called the 1990s Pediatric group) had been added to the study design when the preliminary stereology data from analysis of the first few animals in the 1990s Primate group indicated that they had significant reductions in Purkinje and CA1 cell size compared to controls. The study authors were concerned at the time that any significant findings arising from the stereology analyses of animals in the 1990s Primate group might be attributed to the accelerated schedule in which these animals received TCVs. The inclusion of the 1990s Pediatric group in the study was intended to address those concerns if the stereology findings were consistent across both the 1990s Primate and 1990s Pediatric groups. The advocacy organization agreed to fund the cellular analysis of brain samples on the additional 12 animals in the 1990s pediatric group. The following research plan was approved (taken from the grant): “Year 1 – cut and Nissl stain sections through the cerebellum and hippocampus in 12 male macaque monkeys. Year 2 – using stereology software we will count CA1 hippocampal neurons and cerebellar Purkinje cells and also measure the soma size in the 12 animals. Sections through the hippocampus and cerebellum will be stained for inflammation (glial cell markers) and neurogenesis (double cortin) and densities of these markers will be quantified.” The funding organization provided funding ($25,000) for Year 1 of the grant (Jan 1, 2013 - Dec 31, 2013). During this time, the researchers cut and stained the cerebellar and hippocampal sections for 8 of the 12 animals as outlined in the grant proposal. The cutting and staining of the 4 remaining animals was completed in 2014 when tissues became available. While the funding organization did agree to fund both aims of the study, they did not provide any additional funding in Year 2 to complete the stereology and there was not enough funding available from Year 1 to complete the cellular analysis without additional support.

In the absence of additional funding, as the study progressed and the analysis of the stereology data for the control and 1990s Primate groups was completed, the study authors discussed the need for analysis of the 1990s Pediatric group. This was addressed directly in the paper1: “The neuroanatomical analyses were first performed in brains from the 1990s Primate and 2008 groups, as animals in these groups received the highest amount of EtHg exposure (1990s Primate) or the most extensive vaccine exposure (2008). Because no neuronal differences were found in either of these vaccine groups compared with the control group, no additional vaccine groups were fully studied.”


Why did the early (preliminary) data on the current study that was presented at scientific meetings and as an interim report provided to one of the funding organizations differ from what was published in the final publication?
As is customary for most researchers, preliminary data is often presented at scientific conferences (in both poster and oral format) to elicit feedback from colleagues while the study is still on-going. These interactions not only provide an opportunity for critical review of unpublished work but they often result in new collaborations, and are a very valuable part of research. This study was no exception, and a number of abstracts were presented at various scientific meetings between 2011 and 2015. These data were always identified as preliminary in the published conference abstracts and in the subsequent presentations. To summarize the data presented in these abstracts, the early neuropathology findings from animals in the control and 1990s Primate groups showed a significant increase in Purkinje cell soma area10 and a reduction in Purkinje cell number and hippocampal CA1 cell size.11,12 These findings were based on the cellular analyses of the cerebellum and hippocampus in 4-12 animals per group. The preliminary stereology data presented at conferences between 2011 and 2013 were analyzed by a single, blinded researcher using Stereo Investigator software and a microscope with a 40x objective.  At one of these meetings, a colleague suggested that the study researchers could improve upon the stereology methods being utilized by performing cell counting using a Nanozoomer slide scanning system, which provides 460nm resolution scans of whole tissue slides and allows for magnification of up to 63x, as well as having multiple researchers perform the same analyses to ensure there was no bias. All of the stereology data for each of the regions studied were subsequently analyzed by at least two blinded researchers at two different academic institutions using a Nanozoomer at 63x magnification. Once the final analysis had been completed, the earlier cellular effects were no longer significant.1,13,14 There was no “cherry-picking” of data. Stereology was performed on 16 control animals, 12 animals from the 1990s Primate group, and 8 animals from the 2008 group (4 brains from the 2008 group did not undergo adequate fixation and processing so could not be used in any of the cellular analyses). Behavioral analyses were collected and conducted on all 79 animals, and none of the conference abstracts ever demonstrated significant differences in tests of learning, cognition, and behavior between vaccinated and unvaccinated animals.11-19 Using preliminary data presented at scientific conferences to try to undermine a peer-reviewed, published paper is absurd. 


In a recent news article and blog, a question was raised about earlier findings in both the pilot study and interim reports on this study. Representatives from one of the funding organizations, an advocacy group, reported that they had received interim reports from scientific meeting presentations indicating that there were significant findings. Why weren’t these findings in the final PNAS publication?
In order to address this, it is important to first understand the historical context of the question being raised (see below):


What is a pilot study?
A pilot study is a small-scale preliminary study conducted in order to evaluate feasibility, time, cost, adverse events, and effect size, in an attempt to predict an appropriate sample size and improve upon the study design prior to performance of a full-scale research project.


Where was the pilot study undertaken and how was it designed?
The pilot study was performed at the University of Pittsburgh from 2003-2006. The study was to include 4 control animals that received saline placebo injections and 12 vaccinated animals that followed an accelerated pediatric vaccine schedule based on the vaccines given in the 1990s. This included three thimerosal-containing vaccines and the MMR vaccine. A larger number of animals were assigned to the vaccine group in order to optimize the chances of observing what was anticipated to be an uncommon effect. Animals were raised in an infant nursery and underwent routine blood, urine, and stool collections, assessments of neurodevelopment, cognition, and social behavior, imaging by MRI and PET, surgical collection of GI tissue, and eventual necropsy for tissue collection. The Institutional Animal Care and Use Committee (IACUC) application did not require a power analysis since this was a pilot study. Due to budgetary, space, and staffing constraints as the pilot study progressed, only 17 animals were added to this protocol, 4 controls and 13 vaccinated.


Who funded the pilot study?
This work was supported by the Johnson family, the late Elizabeth Birt, SafeMinds, the Autism Research Institute, the Ted Lindsay Foundation, the Greater Milwaukee Foundation, David and Cindy Emminger, Sandy McInnis, Elyse Roberts, and Vivienne McKelvey.


What were the findings from the pilot study?
There were two published papers arising from the pilot study. The first examined the effect of a single hepatitis B vaccine on the acquisition of neonatal reflexes (involuntary movements or actions such as sucking and grasping) from birth to 14 days of age.5 In this paper, it was reported that the acquisition of three neonatal survival reflexes—root, suck, and snout—was significantly delayed in infant macaques receiving a single thimerosal-containing hepatitis B vaccine at birth. The overall conclusion reported in this paper was as follows: “In summary, this study provides preliminary evidence of abnormal early neurodevelopmental responses in male infant rhesus macaques receiving a single dose of thimerosal-containing hepatitis B vaccine at birth and indicates that further investigation is merited.”

The second paper included preliminary data on structural (MRI) and functional (PET) neuroimaging of the amygdala at approximately 4 and 6 months of age to determine any potential effects of vaccination.6 This analysis included only 2 control animals and 9 vaccinated animals, as these were the only animals from the pilot study with complete imaging data available (i.e. both MRI and PET data from both time-points). The overall conclusion of the study was as follows: “In this pilot study, infant macaques receiving the recommended pediatric vaccine regimen from the 1990’s displayed a different pattern of maturational changes in amygdala volume and differences in amygdala-binding of [11C]DPN following the MMR/DTaP/Hib vaccinations between T1 and T2 compared with non-exposed animals. Because primate testing is an important aspect of pre-clinical vaccine safety assessment prior to approval for human use (Kennedy et al. 1997), the results of this pilot study warrant additional research into the potential impact of an interaction between the MMR and thimerosal-containing vaccines on brain structure and function.”


How do the findings from the pilot study compare to the current study?
The papers from the pilot study did not include any stereology or behavioral data, so they cannot be compared to the newly published PNAS paper1 nor can they be used to contradict the published findings. There is, however, some overlap of data from one of the papers arising from the pilot study and the recent Curtis et al.3 paper: both examined the length of time it took infant primates to acquire neonatal reflexes. In the pilot study, the researchers measured 13 reflexes from birth to 14 days of age in 20 male infants following vaccination with a single thimerosal-containing Hepatitis B vaccine. There were 13 vaccinated animals and 7 control animals (4 infants that received a saline injection and 3 infants that did not receive a saline injection). It should be noted that the 3 infants that did not receive a saline injection were not on the same IACUC protocol but they were raised under similar conditions, and by including these animals during the analysis of data researchers were able to increase the number of animals in the control group. In comparison, the current study examined an additional 6 reflexes (19 reflexes in total) in 6 groups of animals (n=12-16/group) for 20 days, during which time infants received multiple TCVs. It is difficult to directly compare these two studies due to the improved methodology implemented in the Curtis et al. paper. One can easily see that the inclusion of additional reflexes, the assessment of reflexes for a longer period of time, and the inclusion of almost 80 animals is highly likely to produce more robust data. In fact, this issue was highlighted in the Curtis et al. paper: “These data are in contrast to our previous pilot study in which a delay in the acquisition of the root, suck, and snout survival reflexes were reported for primate infants following exposure to the birth dose of the thimerosal-containing Hep B vaccine (Hewitson et al. 2010a). This discrepancy is most likely due to the larger number of animals in the present study providing more accurate estimates.”


Were there any differences in how the current study was implemented compared to the pilot study?
While the studies were conducted at two different research centers, investigators at both facilities implemented a comprehensive battery of neurodevelopmental and cognitive testing, so animals in both studies were raised under very similar conditions. Nursery protocols, developmental assessments, and social behavior testing in both studies were all based on the published protocols from the Infant Primate Research Laboratory at the WaNPRC, although it should be noted that due to staffing issues during the pilot study not all testing was implemented in the same manner.  It should also be noted that the majority of the rhesus macaque infants in both studies were of Indian-origin. In the pilot study, three animals were an Indian-Chinese hybrid (based on shipping and/or breeding records of the dams and sires, although the percentages of Chinese contributions in those animals is unknown). One of these infants was assigned to the control group. In the current study, records indicate that the macaque pregnancies were derived from a breeder colony of Indian-origin macaques for all but 4 hybrid animals: three dams were 1/8 Chinese and one sire was 1/4 Chinese. Two of these infants were assigned to the control group. The use of almost exclusively Indian-origin macaques in this study was important as variability in the genetic make-up of macaques can influence the results of some experiments 7-9 By far the biggest difference between the pilot study and the current study is in the number of animals included.


Is the data from the pilot study of similar caliber to the current study?
No. The pilot study was undertaken in order to examine the feasibility of performing such a complex study in macaques. A number of unanticipated issues with implementation of study procedures for the pilot study were identified – all of which affected the quality of the data collected. For example, while 4 animals had been assigned to the control group, one control animal was mistakenly given a TCV on Day 14, rather than a saline injection, and could not be included in the study (other than the reflex data). A giardiasis infection (not uncommon in animal facilities) occurred in several animal rooms affecting both control and vaccinated animals alike. Seven animals (mostly in the vaccine group) had to be transported to an out-of-state primate facility toward the end of the study, exposing them to many different environmental conditions that had the potential to affect the various blood and tissue samples subsequently collected. In the current study, a power analysis was used to inform study group sizes, improvements to methodologies and study protocols were implemented based on feedback from the pilot study, and additional analyses were undertaken.

 
Why were there only two papers published from the pilot study?
As mentioned above, only 3 of the 4 control animals could be included in the study due to a vaccine being inadvertently given to a control animal in place of a saline injection. This meant that the control group was reduced to two or three animals depending on the dataset. Any knowledgeable scientific researcher understands that a paper with two controls is underpowered, particularly when the experimental group has 5 times as many subjects, and as such, any significant findings should be interpreted with an abundance of caution. This was evident in the conclusions drawn by the study authors in the two papers resulting from the pilot study but unfortunately, some advocacy groups touted these findings as proof that vaccines caused autism, even though the study was not designed to test that hypothesis.

Two other papers from the pilot study were submitted for publication, one examining discrimination learning, and the other examining GI histology. The discrimination learning paper was submitted for publication and reviewed by two experts. The main concerns noted by the reviewers were the very low number of control animals included, inappropriate statistical analyses being implemented, and the lack of additional reversal phases during testing (staffing issues resulted in only one reversal phase being implemented, whereas this type of testing protocol should include multiple reversal testing phases). Since the study authors were unable to address any of the reviewer concerns, this paper was not resubmitted.

The GI histology paper was submitted to a number of different journals. After consistent rejections due to the inclusion of an inadequate number of controls, a study author made the decision to include archived GI tissue from 7 additional rhesus macaques raised at a different primate facility. The data were reanalyzed, and the paper resubmitted to a different journal with the additional control data. One of the reviewers subsequently described the study as “fatally flawed” due to the inclusion of these 7 additional control GI tissue samples. It was also noted that in the original study design “the use of 13 experimental subjects and only 3 controls for a study like this is unacceptable for the detailed analyses that were going to be done, counting various inflammatory cells. An equal number of placebo treated and operated is absolutely required. On Page 8 line 11, the additional incorporation of 7 other controls “raised under similar conditions” is not acceptable. They did not receive the placebo injections nor did they have the opportunity to respond to the pre-surgical interventions, the surgery, and the post-surgical interventions, some of which most certainly would have included antibiotics and corticosteroids. In addition they did not carry with them the embedded sutures which the experimental subjects carried with them, and which may have influenced immune responses.”

After several rejections of these two manuscripts and no ability to address the reviewers’ concerns, the study authors decided to take the knowledge gained from the pilot study and focus their attention on the current larger study that was already underway.


What is the study authors’ response to the recently published critique of the paper posted by an advocacy group who made funding contributions to the project?
Since publication of the first paper (Curtis et al., 2015), the principal investigator and representatives from the organization have contacted the advocacy group and encouraged them to submit any questions directly to the study authors. No response had been received until November 2, 2015. Additionally, the principal investigator contacted this advocacy group prior to the on-line publication of both papers, and will continue to inform them of upcoming publications. It should also be noted that the vast majority of academic journals have strict guidelines regarding the involvement of the funding organization(s) in research. As a matter of transparency, the following statement was included in the manuscript submission to PNAS: “The funding organizations played no role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the presentation, review, or approval of the manuscript.” As per editorial guidelines post acceptance, this statement was not required in the final published version of the PNAS paper.

As for their critique, we are saddened and confused that community activists who claim to seek the truth published what appears to be a review of this paper without asking a single question of any study author, despite being invited to do so. For over ten years a team of dedicated researchers have spent countless hours and made an incredible effort to complete a rigorous and complicated study to answer important questions about vaccine safety. The study authors stand by their findings and publications and will apply the same rigorous and meticulous effort in the analysis and publication of the remaining data sets.


These FAQs will be updated as needed, available here:
http://www.johnson-center.org/research/page/faqs

References:
1. Gadad BS, Li W, Yazdani U, Grady S, Johnson T, Hammond J, Gunn H, Curtis B, English C, Yutuc V, Ferrier C, Sackett GP, Marti CN, Young K, Hewitson L, German DC. Administration of thimerosal-containing vaccines to infant rhesus macaques does not result in autism-like behavior or neuropathology. Proc Natl Acad Sci U S A. 2015 Sep 28. pii: 201500968. Epub ahead of print.
2. Courchesne E, Yeung-Courchesne R, Press GA, Hesselink JR, Jernigan TL. Hypoplasia of cerebellar vermal lobules VI and VII in autism. N Engl J Med. 1988 May 26;318(21):1349-54.
3. Curtis B, Liberato N, Rulien M, Morrisroe K, Kenney C, Yutuc V, Ferrier C, Marti CN, Mandell D, Burbacher TM, Sackett GP, Hewitson L. Examination of the safety of pediatric vaccine schedules in a non-human primate model: assessments of neurodevelopment, learning, and social behavior. Environ Health Perspect. 2015 Jun;123(6):579-89. doi: 12.1289/ehp.1408257. Epub 2015 Feb 18.
4. Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum Associates.



5. Hewitson L, Houser LA, Stott C, Sackett G, Tomko JL, Atwood D, Blue L, White ER. Delayed acquisition of neonatal reflexes in newborn primates receiving a thimerosal-containing hepatitis B vaccine: influence of gestational age and birth weight. J Toxicol Environ Health A. 2010;73(19):1298-313. doi: 10.1080/15287394.2010.484709.
6. Hewitson L, Lopresti BJ, Stott C, Mason NS, Tomko J. Influence of pediatric vaccines on amygdala growth and opioid ligand binding in rhesus macaque infants: A pilot study. Acta Neurobiol Exp (Wars). 2010;70(2):147-64.
7. Smith DG, McDonough J. Mitochondrial DNA variation in Chinese and Indian rhesus macaques (Macaca mulatta). Am J Primatol. 2005 Jan;65(1):1-25.
8. Malhi RS1, Sickler B, Lin D, Satkoski J, Tito RY, George D, Kanthaswamy S, Smith DG. MamuSNP: a resource for Rhesus Macaque (Macaca mulatta) genomics. PLoS One. 2007 May 9;2(5):e438.
9. Jiang J, Kanthaswamy S, Capitanio JP. Degree of Chinese ancestry affects behavioral characteristics of infant rhesus macaques (Macaca mulatta). J Med Primatol. 2013 Feb;42(1):20-7. doi: 10.1111/jmp.12026. Epub 2012 Nov 29.
10. Shrikanth B, Li W, Hewitson L, Ferrier C and German DC. (2011) Influence of pediatric vaccines on CNS development in the rhesus macaque: Cerebellar Purkinje cells Program No. 57.02. Society for Neuroscience, 2011, Washington, DC.
11. Gadad B, Li W, Curtis B, Yutuc V, Ferrier C, Sackett G, Young KA, Sachsenmaier S, Hewitson L and German DC. (2012) Influence of pediatric vaccines on CNS development and behavior in the rhesus macaque. Program No. 443.12. Society for Neuroscience, 2012. New Orleans, LA.
12. Gadad B, Li W, Grady S, Yazdani U, Curtis B, Yutuc V, Ferrier C, Sackett G, Hewitson L and German DC. (2013) Influence of pediatric vaccines on CNS development and behavior in the rhesus monkey. Program No. 418.14. Society for Neuroscience, 2013. San Diego, CA.
13. Gadad B, Li W, Yazdani U, Curtis B, Yutuc V, Ferrier C, Sackett G, Young KA, Darden JA, Hewitson L and German DC. (2014) Influence of pediatric vaccines on CNS development and behavior in the rhesus macaque: Relevance to autism. Program No. 799.05. Society for Neuroscience, 2014. Washington, DC.
14. Hewitson L, Gadad B, Li W, Grady S, Curtis B, Yutuc V, Ferrier C, Sackett G, and German DC. (2015) Does administration of thimerosal-containing vaccines to infant rhesus macaques result in an autism-like neuropathology? Neurotoxicology and Teratology 49:NTX106.
15. Curtis B, Morrisroe K, Yutuc V, Ferrier C, Sackett GP and Hewitson L. (2014) Influence of Pediatric Vaccines on Social Behavior in the Rhesus Monkey. Neurotoxicology and Teratology 43:NBTS P02.
16. Holden C, Sharma N, Kenney C, Curtis B, Linerato N, Rulien M, Ferrier C, Marti CN, Sackett GP and Hewitson L. (2014) Neonatal Reflexes in Infant Macaques (Macaca mulatta) Exposed to Low-Dose Thimerosal via Vaccination. Neurotoxicology and Teratology 43:NBTS P03.
17. Hewitson L, Curtis B, Liberato N, Rulien M, Ferrier C, Marti CN, Mandell D and Sackett, GP. (2014) Learning and Cognition in Infant Macaques (Macaca mulatta) Exposed to Low-Dose Thimerosal via Vaccination Neurotoxicology and Teratology 43:NBTS 23.
18. Hewitson L, Curtis B, Liberato N, Kenney C, Yutuc V, Ferrier C, Marti CN and Sackett, GP (2015). Assessment of social behavior in non-human primate infants following administration of thimerosal-containing vaccines. #19052. International Meeting for Autism Research (IMFAR) May 13 - 16, 2015, Salt Lake City, UT.
19. Hewitson L, Curtis B, Yutuc V, Ferrier C, Marti CN, and Sackett G (2015). Social behavior in non-human primate infants and juveniles following administration of thimerosal-containing vaccines. Neurotoxicology and Teratology 49:NTX137.

 

FREQUENTLY ASKED QUESTIONS ON PREVIOUS STUDIES

 

Examination of the Safety of Pediatric Vaccine Schedules in a Non-Human Primate Model: Assessments of Neurodevelopment, Learning, and Social Behavior. B. Curtis et al. Environmental Health Perspectives, Feb 18, 2015
http://ehp.niehs.nih.gov/1408257

Updated September, 2015

The following statement was submitted as a comment on PubMed Commons:

    Supplementary Figure 5 clearly shows a drastic reduction in learning in the thimerosal-exposed group. The authors discussion: “In the present study animals in the TCV group appeared to perform poorer than controls in learning set testing but showed little evidence that their responses had organized into a strategy that was different from that of the control group. In fact, the reported difference was only found in the overall mean averaged across all of the blocks and trials, not in their learning across trials or blocks, which is the outcome needed to indicate a strategy difference.” But in fact, a deficit inclearning seems to be in multiple groups, for if one looks at group E, there seems to be a slope difference from the control signifying a key difference between exposures for learning strategy. These results are not reported. Perhaps Supplemental Figure 5 results should have been the title of this study instead: “Ethylmercury from vaccines reduces learning capacity.”

Response to Comment:

It appears that the commenter is confusing choice reaction time (latency) with learning. In Supplementary Figure 5, reaction time is scored as the time it takes for an animal to respond to the presented problem. Reaction times can increase for a variety of reasons making it difficult to interpret this at a group level. One reason reaction times can increase is because an animal is bored and “checks-out” of the task. For example, in discrimination learning, when the rewarded object color is first reversed, reaction times and balking behavior increase dramatically. Learning set, a more complex discrimination test, is particularly difficult in macaques of this age, and as such, animals will slow down when they are no longer motivated to participate. This was evident in Supplementary Figure 3, which shows that none of the groups, including the controls, developed an organized learning set.

To specifically address whether thimerosal was affecting reaction times in Learning Set testing, one only has to compare reaction times for animals in the TCV and 1990’s Primate groups, which both received the exact same dosing and timing of thimerosal, to see that there was no consistent effect. As reported in the paper (p. 584), only the TCV and 2008 groups showed a significant difference in reaction times compared to the controls, not animals in the 1990s Primate group.

Q: Will the dataset be made available to other researchers?

A: We are in the process of creating MTAs so that we can license the various datasets and biological samples to other researchers. These will be made available after completion of publication of papers arising from this study.

FREQUENTLY ASKED QUESTIONS - POSTED FEBRUARY, 2015


Why was this study done?
In light of concerns reported by some parents and advocacy groups over the safety of pediatric vaccines, we decided to examine the safety of a variety of vaccine schedules, including the current pediatric vaccine schedule, which has greatly expanded over the last decade.


What was this study investigating?
This study was designed to examine the safety of US-recommended pediatric vaccine schedules in non-human primates (rhesus macaques). Different groups of animals received vaccines like those used in the 1990s; the measles-mumps-rubella (MMR) vaccine; or vaccines like those used in the current recommended US vaccine schedule. Several of the 1990s vaccines contained thimerosal, whereas most current vaccines do not. (The MMR vaccine has never contained thimerosal.) Vaccinated and unvaccinated infant primates underwent extensive neurodevelopmental, cognitive, and behavioral testing from birth to 12 months of age.


What is Thimerosal?
Thimerosal is an ethyl mercury-containing organic compound. It is used as a preservative in a number of biological and drug products to help prevent contamination from harmful microbes. It has been used in many multi-dose pediatric vaccines, including Hepatitis B, Diphtheria-Tetanus-acellular Pertussis (DTaP), and Haemophilus influenzae B (Hib), as well as some flu vaccines. Thimerosal has now been removed from most vaccines recommended for pediatric use in the US, though it is still used as a preservative in some flu and meningitis vaccines.


What were the findings of this study?
Perhaps the most significant finding, with respect to the hypothesis that thimerosal-containing vaccines may have an adverse effect on behavior, was that all animals in the study developed the expected repertoire of social behaviors typically found in rhesus macaque infants of this age. Furthermore, virtually no negative behaviors, such as rocking, self-clasping, and stereotypy (repetitive behaviors), were reported, regardless of vaccination status. The very few instances of negative behaviors observed in the primates were not definitively associated with being vaccinated, and in some assessments vaccinated animals performed better than controls.


Does this answer the question of whether vaccines cause autism?
This non-human primate study of vaccine safety was not designed to determine whether vaccines cause autism, nor can the data be used to refute that idea. However, the study did demonstrate that there was no evidence of impaired neurodevelopment or aberrant social behavior following vaccination. In light of the concern that vaccines may be associated with autism, the results from this paper are encouraging since social development in infant primates was normal.

It should be noted that all of the animals included in this study had a normal birth weight and gestational age. Low birth weight and decreased gestational age have been associated with increased autism risk. These and other variables, such as gender and genetic factors, were not addressed by this study. Furthermore, there have been a number of studies that suggest that environmental factors during pregnancy may contribute to autism risk. It should also be noted that there are many different health conditions clinically associated with autism that may possibly make the fetus or infant more vulnerable to a number of environmental factors. These factors were not accounted for in this study.


Why did you assess behavior in animals only until they were 12 months of age?
It has been well established that rhesus macaque infants develop approximately four times faster than human infants, particularly in terms of the infant visual system, pattern recognition, and the acquisition of object concept permanence, all of which were tested in this study. The vaccine-dosing schedule was therefore adjusted to accommodate this projected 4:1 developmental timeline for infant primates. At 12 months of age, a rhesus macaque would be considered developmentally similar to a 4-year-old child, an age at which autism should be easily identified in affected children.


Since this study took over five years to complete, how could you ensure that all behavioral technicians performed the assessments the same way?
Care was taken to ensure all behavioral testers remained blinded to study group assignments, and they were fully trained to the highest standard. Furthermore, the assessments followed very detailed protocols that have been used at the Infant Primate Research Laboratory for over three decades.


What other papers will be published based on this study?
This was a comprehensive five-year study examining aspects of neurodevelopment, growth measures, behavior, metabolic and immune markers, brain structure and function, and neuropathology.  We expect to publish a number of papers over the next year and a half examining the safety of pediatric vaccine schedules in infant primates.


Who were the study co-authors?
- Principal Investigator: Laura Hewitson, PhD, Director of Research, The Johnson Center for Child Health & Development; Adjunct Associate Professor, Department of Psychiatry, University of Texas Southwestern, Dallas TX; Affiliate Scientist, Washington National Primate Research Center, Seattle WA.
- Gene Sackett, PhD, Professor Emeritus of Psychology, University of Washington, Seattle WA; Infant Primate Research Laboratory, Washington National Primate Research Center and Center on Human Development and Disability, University of Washington, Seattle.
- Thomas Burbacher, PhD, Professor of Environmental and Occupational Health Sciences, University of Washington, Infant Primate Research Laboratory, Washington National Primate Research Center and Center on Human Development and Disability, University of Washington, Seattle, WA.
- C. Nathan Marti, PhD, Independent Consultant, Abacist Analytics, LLC, Austin, TX
- Dorothy Mandell, PhD, Independent Consultant, Austin, TX
- Britni Curtis, B.A. Infant Primate Research Laboratory, Washington National Primate Research Center, Seattle, WA.
- Noelle Liberato, B.A. Infant Primate Research Laboratory, Washington National Primate Research Center, Seattle, WA.
- Megan Rulien, B.S. Infant Primate Research Laboratory, Washington National Primate Research Center, Seattle, WA.
- Kelly Morrisroe, B.S. Infant Primate Research Laboratory, Washington National Primate Research Center, Seattle, WA.
- Caroline Kenney, B.S. Infant Primate Research Laboratory, Washington National Primate Research Center, Seattle, WA.
- Vernon Yutuc, B.A. Infant Primate Research Laboratory, Washington National Primate Research Center, Seattle, WA.
- Clayton Ferrier, B.A. Infant Primate Research Laboratory, Washington National Primate Research Center, Seattle, WA.


Who else was involved in the study?
There were a number of staff members at the Infant Primate Research Laboratory at the Washington National Primate Research Center who were involved in this study, as well as staff at the California National Primate Research Center, and the Environmental Research Training Laboratory at the University of Kentucky.


Who funded this study?
This work was supported by The Ted Lindsay Foundation, SafeMinds, National Autism Association, the Vernick family, and the Johnson family. This work was also supported by WaNPRC Core Grant RR0166 and CHDD Core Grant HD02274. None of the funders were involved in any aspect of the study, including the study design, implementation of the study, data analysis, the decision to publish, or the review of manuscript drafts prior to acceptance for publication.


What are the differences in this study and the pilot study that was published in 2010?
The two studies are very different. In 2010, Dr. Hewitson published a pilot study in the Journal of Toxicology and Environmental Health, in which delays in the acquisition of several survival reflexes were reported following a single birth dose of Hepatitis B vaccine.  This is in contrast to the data reported in this study, perhaps due to the much larger number of animals in the present study providing more accurate estimates. Additionally, in the current study, 6 additional reflexes were tested (a total of 19 reflexes), and not only after the birth dose of Hepatitis B, but also after multiple thimerosal-containing vaccines had been administered during the first 21 days, providing a longer-term view of any possible effects. While the studies were conducted at two different research centers, investigators at both facilities implemented a comprehensive battery of neurodevelopmental and cognitive testing, so animals in both studies were raised under very similar conditions. It should also be noted that the majority of the infants in both studies were of similar genetic make-up in regards to Indian versus Chinese origin. This is important to note as variability in the genetic make-up of macaques used in a study can influence the results on those experiments. There were several collaborators and colleagues involved in the conception and implementation of the pilot study who were not involved in the current study, including Lisa Houser, Dr. David Atwood, Dr. Carol Stott, the late Gerald Ruppenthal, Dr. Saverio Capuano, Dr. Andrew Wakefield, Dr. Amanda Dettmer, Dr. Lisa Blue, and Elizabeth Railey White. 


Click here to see some of our current research studies.

Click here to see some of our recently published studies.