Академический Документы
Профессиональный Документы
Культура Документы
4th Regional
Pneumococcal
Symposium
March 2–3, 2009 | Johannesburg, South Africa
PneumoADIP
GlaxoSmithKline
TABLE of contents
Foreword i
Introduction i
Executive Summary ii
Session III: Impact of Pneumococcal Conjugate Vaccines — Direct and Indirect Effects 19
Speakers 53
ATTENDEES 54
FOREWORD
The Fourth Regional Pneumococcal Symposium, held in Johannesburg, South Africa, in March 2009, followed successful
symposia in Mexico, Brazil and Turkey. The symposia convened in 2004, 2006, and 2008 brought together hundreds of partici-
pants to discuss the substantial progress made toward safe, effective pneumococcal vaccines, and to find ways to bring
those vaccines to the world’s poorest children. Since the first symposium in 2004, major advances have been made in
quantifying the burden of pneumococcus, raising awareness about the need for immunization programs and developing the
vaccines themselves.
After focusing on the Americas and Central Asia, the fourth symposium presented new data highlighting the specific impact
of pneumococcal disease in sub-Saharan Africa, including co-morbidity with other conditions highly prevalent in Africa
such as HIV and sickle cell disease, and reviewed lessons learned from vaccine trials on the continent. As in each of the
earlier symposia, presenters and participants sought to strengthen both regional and worldwide efforts to prevent the
disease through immunization programs, discussing new vaccines in development, and addressing the scientific, social, and
economic challenges associated with bringing the vaccines to the poorest parts of the world.
We are grateful to all the researchers, doctors, epidemiologists, public health workers, economists, members of industry,
policy makers, donors, and others whose efforts made this Fourth Regional Pneumococcal Symposium possible, and who will
have a tremendous impact on the future of Africa’s children.
INTRODUCTION
The Fourth Regional Pneumococcal Symposium was held 2–3 March 2009 in Johannesburg, South Africa, bringing together
more than 185 participants from 38 countries to discuss the current global impact of pneumococcal disease, celebrate prog-
ress made and commit to continued progress toward its prevention.
“Our efforts continue to raise global awareness of the need for a vaccine to prevent pneumococcus, one of the leading killers
of children in the poorest parts of the world, and we are on our way toward implementing the vaccine worldwide,” said Dr.
Ciro de Quadros of the Sabin Vaccine Institute. “I believe that with the introduction of pneumococcal and rotavirus vaccines
in the world today, we can cut by more than half the mortality of children under five.”
This fourth pneumococcal symposium marks a significant shift in policy and science from the first symposium, held in 2004.
“I think it has been amazing to see the progress from where we were just five years ago, to what one can observe today in
terms of pneumococcal vaccination,” said Dr. Jean-Marie Okwo-Bele of the World Health Organization (WHO). “The awareness,
the supply issues, even the level of financing has improved.”
“I don’t think there has been a more exciting time in terms of the opportunities for prevention. Although there are still
barriers to overcome in terms of vaccine introduction, we now have new funding opportunities and new products available
that have overcome some of those most important barriers,” said Dr. Cynthia Whitney of the U.S. Centers for Disease Control
and Prevention.
Building and sustaining that political will requires action from around the globe, and the symposium sought to rally
researchers, public health workers, doctors, vaccine developers and manufacturers, leaders of industry, and policy makers
to action, framing pneumococcal disease prevention not just as a public health issue, but as a moral responsibility.
EXECUTIVE SUMMARY
Despite recent advances in vaccine development, pneumococcal disease is still a serious threat to life and health across the
developing world, particularly among children. Current estimates put the death toll from pneumococcal disease at 820,000
worldwide each year, with more than half of those in Africa. Many millions more suffer from pneumonia, middle ear infection,
blood poisoning, and meningitis.
Of the 14 million annual cases of pneumococcal disease worldwide, 4 million are in Africa, giving the continent a higher-than-
average mortality rate from the disease. The Fourth Regional Pneumococcal Symposium, held in Johannesburg, South Africa,
on 2–3 March 2009, focused attention on the burden of pneumococcal disease in Africa and around the world, looking for the
best ways to prevent and address the disease among the world’s poorest children.
More than 180 people from 38 countries, including 23 African countries, attended the Fourth Regional Pneumococcal Sympo-
sium in Johannesburg, South Africa. During the symposium, participants heard presentations and discussed topics ranging
from pneumococcal epidemiology to country perspectives on vaccine introduction and disease surveillance.
Following successful symposia in Mexico City, Mexico, in 2004; São Paulo, Brazil, in 2006; and Istanbul, Turkey, in 2008; the
meeting was organized by the Sabin Vaccine Institute and PneumoADIP in partnership with the World Health Organization
(WHO), the South African Paediatric Association, the U.S. Centers for Disease Control and Prevention (CDC), the Kenya Paedi-
atric Association, and the University of Witwatersrand. Since the first symposium in 2004, major advances have been made
in quantifying the burden of pneumococcus, raising awareness about the need for immunization programs, and developing
and introducing the vaccines themselves.
After focusing on the Americas and Central Asia, the fourth symposium presented new data highlighting the specific impact
of pneumococcal disease in sub-Saharan Africa, including co-morbidity with other conditions highly prevalent in Africa, such
as HIV and sickle cell disease, and reviewed lessons learned from vaccine trials on the continent. As in each of the earlier
symposia, presenters and participants sought to strengthen both regional and worldwide efforts to prevent the disease
through immunization programs, discussing new vaccines in development and addressing the scientific, social, and economic
challenges associated with bringing the vaccines to the poorest parts of the world.
Helping to kick off the meeting, South African Deputy Minister of Health Dr. Molefi Sefularo discussed the importance of pneu-
mococcal disease in South Africa and spoke about the country’s recent introduction of the 7-valent pneumococcal conjugate
vaccine (PCV7). “Successful immunization programs have resulted in a decrease in incidence and prevalence of diseases
like polio and the complete eradication of diseases like smallpox,” Dr. Sefularo said. “We believe that the introduction of the
pneumococcal vaccine will reduce mortality amongst children and the elderly.”
Another strong benefit imparted by PCV is the potential to achieve Millennium Development Goal 4 to reduce child mortality.
“By increasing the priority to prevent pneumonia, meningitis, and pneumococcal disease, African countries can make substan-
tial progress towards reaching the Millennium Development Goals for improved child survival and health,” said Orin Levine,
Executive Director of PneumoADIP. “The next step is really action at the country level, and that action has already begun.”
“African children have very high rates of child mortality, and pneumococcal disease is a big contributor,” said Dr. Levine.
“They have the highest rates of pneumococcal disease in the world, and mortality rates from it are typically 10 to 50 times
in Africa what they are in the U.S. and Europe.”
Among the presentations with new data were two studies conducted by the Sabin Vaccine Institute, the Pneumococcal Aware-
ness Council of Experts (PACE), and PneumoADIP, with results indicating a high burden of bacterial meningitis (pneumococcal
meningitis neurological sequelae) and an increased risk of invasive pneumococcal disease for people with sickle cell disease
in African populations. There were also presentations looking at serotype distribution and replacement disease as well as the
high burden of pneumococcal disease and the benefits of PCV in HIV-infected individuals.
“Especially in Africa, the HIV epidemic emphasizes the need for pneumococcal vaccination and other vaccines, because HIV-
infected children are at significantly higher risk of acquiring pneumococcal disease than children who are not infected,” said
Till Bärnighausen, Senior Epidemiologist at the Africa Centre for Health and Population Studies and Associate Professor of
Health and Population Studies at the University of KwaZulu-Natal, South Africa. “They are nine to 43 times more likely than
non-HIV-infected children to develop pneumococcal bacteremia.”
Some of the key research findings presented at the symposium and reflecting the burden of pneumococcal disease in Africa
include:
• Sickle cell disease increases the risk of pneumococcal disease by 37 times, and together with HIV infection, malnutri-
tion, and indoor air pollution, it puts African children at high risk
• Even when treated with antibiotics in a hospital, up to one-half of all children in Africa who get pneumococcal menin-
gitis will either die or be disabled as a consequence of the disease
• Of approximately 14 million annual cases of pneumococcal disease worldwide, 4 million (28 percent) are in Africa
• Of approximately 820,000 deaths due to pneumococcal disease worldwide, 440,000 (49 percent) are in Africa
• About 90,000 of pneumococcal deaths worldwide are HIV-positive children
• Eighty-nine percent of pneumococcal deaths are from pneumonia, 7 percent from meningitis, and 4 percent from NPNM
• Pneumococcal conjugate vaccines (PCV) reduced under five mortality by 16 percent in rural African settings
• P CV reduced hospitalizations by 15 percent in rural African settings and prevents the long-term costs of caring for
disabled survivors
Since earlier symposia were convened, pneumococcal vaccines have been introduced in some countries in Africa, with many
valuable lessons learned. Country representatives from South Africa and Gambia candidly shared their previous experiences
with vaccine introduction, including some of the hurdles that remain for PCV introduction in their countries. According to the
speakers, some of the bigger challenges that remain are affordability and having a sustainable financing plan, developing the
infrastructure and capacity to safely and properly handle the vaccine, generating the political will to implement PCV immu-
nization programs, and developing coordinated immunization campaigns. In addition to the African country perspective, Jim
Dobbin, a member of the British Parliament, spoke about his work as the vice-chair of the All-Party Parliamentary Group on
Pneumococcal and the role of donor countries.
Cost-effectiveness is also a key issue, as advocates seek to expand PCV immunizations to additional countries. Looking at
the potential cost savings that come from the herd immunity effect of immunizations, and at the reduced hospitalization
costs resulting from fewer cases of the illnesses, presenters argued that PCV is cost-effective and in some cases can actually
result in cost savings. “There is a profound effect on cost savings when one incorporates herd protection,” said Dr. Anusha
Sinha, of the New Jersey Medical School. “The vast majority of studies [of PCV7’s direct effects] have found this intervention
to be cost-effective.”
Continued Action
Wrapping up a great symposium, the final session was focused on moving forward and continuing efforts to advocate for the
prevention of pneumococcal disease. Orin Levine, co-chair of the symposium, described the meeting as a “call for continuing
action.” A member of the organizing committee, Fred Were of the Kenya Paediatric Association, prompted his colleagues to
continue working hard to “get the map [of PCV introduction status] the same color.”
While conference attendees are optimistic about the impact of the steps taken thus far, challenges remain. Shabir A. Madhi,
Co-Director of the Respiratory and Meningeal Pathogens Research Unit at the University of Witwatersrand, said, “The chal-
lenge for us now over the course of the next five to ten years is to lay the platform to advocate for the use of this vaccine in
other parts of Africa where the need for this vaccine is the greatest.”
Dr. Ciro de Quadros summed up the symposium: “Africa can do it, but support is needed.”
During this session, Dr. Orin Levine, Executive Director of PneumoADIP, PACE Co-Chair, and Associate Professor at Johns Hopkins
Bloomberg School of Public Health, reported on the urgency to prevent pneumococcal disease and pneumonia in African chil-
dren and the success of pneumococcal vaccines in Africa thus far.
Dr. Levine also highlighted the role pneumococcal vaccines can play in helping reach Millennium Development Goals, espe-
cially those aimed at reducing poverty, reducing child mortality, and even improving educational outcomes.
Pneumococcal Deaths in
Children aged 1-59 mos per
100,000 population of
children under age 5
< 10
10 - 100
100 - 300
300 - 500
Over 500
In the four years since the Gambia trial results were published, the World Health Organization has turned that evidence into
a policy recommendation for the introduction of PCV in all countries. It recognizes the burden of pneumococcal disease and
the safety and efficacy of PCV, and urges pneumococcal prevention as a priority, particularly for those countries with high
infant and child mortality.
Cost-Effectiveness
The pneumococcal conjugate vaccine (PCV) prevents substantial morbidity and mortality, and meets criteria of being “highly
cost-effective” at prices of $5 per dose or less in almost every GAVI country. Dr. Anushua Sinha published a paper in The
Lancet in February 2007 that examined the cost-effectiveness of PCV in the prevention of child mortality, and found in this
analysis that at a vaccine cost between $1 and $5 per dose, purchase and accelerated uptake of PCV in the world’s poorest
countries is projected to substantially reduce childhood mortality and to be highly cost-effective.
Vaccine Outlook
The 7-valent pneumococcal conjugate vaccine (PCV7) is moving toward more widespread introduction. The vaccine is safe,
effective, and can be given in existing schedules. The manufacturer, Wyeth, is willing to immediately supply it at tiered
prices for GAVI-eligible countries, including Rwanda and the Gambia. The next pneumococcal conjugate vaccine, the 10-valent
pneumococcal Haemophilus influenzae protein D conjugate formulation (PHiD-CV), has recently been licensed in Canada and
is expected to be released elsewhere. In 2010, a 13-valent formulation (PCV13) is expected to replace PCV7. The outlook for
supply and the types of vaccines needed to maximize child survival impact is better than ever.
The United States, Australia, New Zealand, and high-income countries in Western Europe are providing these vaccines to
their children routinely, and the next step is to introduce the vaccines to low-income countries with the highest risk of pneu-
mococcal disease. Africa currently has the most countries who have expressed interest to GAVI in introducing the vaccine.
Although significant progress has been made in vaccine development, policy, and financing, there are still hurdles ahead.
The progress that has been made is part of an integrated package of child survival tools, and as these are delivered, it is
important to remember the bacteria will try to stay ahead of the vaccine technology. Research must continue to generate
new vaccines that outpace bacteria development. Sustaining political will is vital to ensuring that awareness of the disease
burden and the potential vaccine impact and its benefits are well-known.
$22 per
DALY*
$450 $1350
Below per capita GDP 1 — 3x per capita GDP Over 3x per capita GDP
* ASSUMPTIONS:
• Model developed prior to availability of GDB estimates
• Deaths and cases prevented extrapolated from Gambian trial vaccine efficacy
• Direct and indirect protection in children less than 5 years of age
• Three doses of vaccine at $5 per dose
This session focused on the epidemiology and burden of pneumococcal disease with a concentration on invasive pneumo-
coccal disease and pneumonia in Africa. Dr. Thomas Cherian, Coordinator of the Expanded Program on Immunization at the
World Health Organization (WHO) Department of Immunization, Vaccines and Biologicals opened the session with an overview
of the incidence of pneumococcal invasive disease in Africa and outlined the three models he and his research team used to
estimate the burden of disease.
Dr. Hope Johnson from PneumoADIP reported on the results from the Pneumococcal Global Serotype Project. Dr. Shabir Madhi
of the University of the Witwatersrand discussed the importance of examining pneumococcal disease and HIV together, specifi-
cally among children under five and adults aged 24–44. Dr. Fred Were from the Kenya Paediatric Association described the
burden of pneumococcal meningitis and sequelae in Africa, and illustrated the burden of care that falls upon families and
societies that have afflicted children. Finally, Dr. Meena Ramakrishnan presented the second of two literature reviews that
highlighted data from Africa on bacterial infections in persons with sickle cell disease, with a specific focus on invasive pneu-
mococcal disease.
1000
458
Incidence, cases/100,000 pop
166.9 213
100
45.3 49.1
10
1
Finland
Chile
USA,
multisite
Kenya
Gambia
Not only is the disease incidence higher in Africa, the severity and case fatality from these diseases is also higher
Sources: Robinson KA JAMA 2001; Davidson M JID 1994; O’Dempsey TJ PIDJ 1996; Levine MM PIDJ 1998; Eskola J JAMA 1992; Berkley NEJM 2005
Pneumococcal disease has a bimodal age distribution, so in the United States, there is a high disease incidence in those
less than two years of age and in the elderly population. There are similar disease patterns in Africa, with data from Kenya
showing high rates of disease in infants that goes down slightly but still remains substantially high in those one to two years
of age, and then declines beyond the age of five years.
“Surveillance doesn’t pick up the entire burden of disease, but is very important
because it provides empirical data on disease etiology and epidemiology and also
allows researchers to monitor changes following the introduction of an intervention.”
— T h o m a s C h er ia n
Although important, surveillance has its shortcomings. Many surveillance studies are facility-based, which can raise ques-
tions about the quality of the data. Case ascertainment and access to care are limited in developing countries, particularly
in Africa, so facility-based surveillance often results in missed cases.
Laboratory performance also affects data quality, but even in good laboratories, prior use of antibiotics often limits the
accuracy of surveillance that is based on laboratory confirmation through culture. Ultimately, this leads to an underestima-
tion of the burden of disease. The sensitivity of culture methods for diagnosing pneumococcal disease is particularly acute
for pneumonia, because very few cases of pneumococcal pneumonia have a positive culture.
Overcoming these barriers to quantify the burden of pneumococcal disease is necessary to determining whether an inter-
vention is cost-effective. One approach is through the use of mathematical modeling. Researchers are able to restrict the
data to those that meet quality indicators and are representative of the populations in which they were conducted.
In this case, researchers sought to estimate the number of cases and number of deaths with a model that would allow them
to predict the cases of Hib and pneumococcal meningitis, pneumonia, and other invasive diseases averted through vaccina-
tion. As a requirement for WHO, the researchers had to transparently document the data sources, methods, and assumptions.
Also, through this process, the researchers are expected to make future recommendations on how to improve the precision
and robustness of data collection.
Dr. Cherian described the breakdown of the three models used in the study, which included one for meningitis, one for inva-
sive disease other than pneumonia and meningitis, and one for pneumonia.
The systematic literature review included nine databases and produced more than 15,000 citations. A team of screeners and
abstracters pared the search down to about 900 papers that were then subjected to inclusion and exclusion criteria. Abstrac-
tion was performed on a total of 336 studies which were included in the analysis database.
In instances where the researchers did not have data for a particular country, the world was divided according to
World Bank regions and stratified by under five mortality rates of low, medium, high, and very high. The researchers
assumed that disease epidemiology and burden were similar within geographic regions and within those regions by
mortality strata. So, within a region or sub-region, the countries that fall in the same mortality strata would have a
similar disease epidemiology and burden.
Model 2
In the second model, non-pneumonia, non-meningitis invasive disease incidence was determined by taking the ratio
of meningitis to non-pneumonia and non-meningitis cases. This was replicated across several studies using meta-
analysis to produce one ratio of meningitis to non-meningitis invasive disease. The meningitis cases in each country
were then multiplied with this ratio to get the number of non-pneumonia, non-meningitis cases. A similar process was
used with the case fatality ratio to come up with the number of deaths.
Model 3
Using data from vaccine clinical trials, the researchers estimated that there were about 14 million cases of pneumo-
coccal disease around the world, with 4 million occurring in Africa; there were about 800,000 pneumococcal deaths,
of which 90,000 were among people who were HIV-positive; and there were 440,000 pneumococcal deaths in Africa
specifically, which accounted for 49 percent of global deaths. Most of these deaths were caused by pneumonia, with
only a small proportion attributed to meningitis and other invasive diseases.
Death rates have declined in South Asia, but are higher in West Africa and Central Africa because of the higher case
fatality ratio. In many African countries, the incidence of deaths is over 300 per 100,000 children under the age of five
— and of the 10 countries with the highest rates of death for this age group, five countries are in Africa.
The goal of the project was to support evidence-based decisions on development and introduction of serotype-based
vaccines. WHO asked the research team to provide a serotype analysis to the Target Product Profile Committee, which was
charged with establishing the minimum characteristics of vaccines to be eligible for the Pneumococcal AMC.
The research team aimed to identify and collect all existing data on serotypes causing invasive pneumococcal disease in
children worldwide, to describe the serotype distribution globally and by region, and to determine what variables (such as
age and syndrome) influence serotype distribution.
Methodology
The researchers conducted a literature review identifying citations from four existing literature reviews; systematically
searched 14 literature databases for studies with pre-vaccine introduction serotype data published between 1980 and June
of 2007; and obtained unpublished data from researchers worldwide conducting surveillance on invasive pneumococcal
disease. The researchers identified 1,230 articles and 60 datasets that were reviewed for inclusion in the analysis, and then
conducted a meta-analysis to establish the proportion of disease caused by each serotype using a random effects model.
Globally, serotypes 14, 6B, and 1 are the most commonly occurring serotypes; and some 10
serotypes account for 70 percent of invasive pneumococcal disease (range: 9-12 serotypes).
Next, the researchers estimated serotype distributions by categorizing countries according to United Nations regions. They
focused their analysis on Africa and Asia because this is where pneumococcal disease incidence and mortality is the highest
and also because in previous analyses, Africa and Asia had very different serotype coverage from each other.
Six serotypes in North America account for 70 percent of disease in the country, with nine serotypes in Africa and 11 sero-
types in Asia needed to comprise the same percentage. In Asia, serotypes 14 and 6B were co-dominant and five serotypes
accounted for 50 percent of disease.
In Africa, serotype 14 was also the most common, followed by serotypes 1 and 5. Interestingly, the researchers found that
the seven most commonly occurring serotypes in Africa were identical to those in Asia, although the proportion of disease
caused by the specific serotypes varied between the two regions. The researchers identified a limited set of seven serotypes
(1, 5, 6A, 6B, 14, 19F and 23F) that accounted for 58–66 percent of invasive pneumococcal disease across regions.
Next, the researchers estimated the potential impact of various vaccine formulations on pneumococcal disease. Regional
invasive pneumococcal disease was represented by serotype distribution and vaccine formulations, and it showed that
vaccine serotypes included in PCV7 account for approximately 50 percent or greater of total invasive pneumococcal disease
in children under five across all regions. To account for 70 percent of disease would require serotypes that are included in
the 10- or 13-valent vaccines.
To estimate the impact various vaccine formulations would have on pneumococcal disease, the researchers used the pneumo-
coccal disease burden data from the WHO Pneumococcal Global Burden of Disease (GBD) project. This data indicates that inci-
dence rates of pneumococcal disease due to vaccine serotypes are highest in Africa and Asia for all three vaccine products.
Looking at global pneumococcal deaths by vaccine serotypes shows that serotypes included in PCV7 could help prevent
approximately 400,000 pneumococcal deaths in children each year; vaccine serotypes account for approximately an addi-
tional 170,000 deaths in PCV10, and 200,000 in PCV13.
Data Limitations
Throughout the course of this project, researchers found that published literature has features that limit analysis and inter-
pretation of serotype data. In some cases, inconsistent or incomplete data on individual serotypes is presented, and in
other cases only serogroup data is presented — “pooled” by serotypes into PCV formulations — or many individual serotypes
lumped into “other.” The researchers also found it difficult to interpret “missing” data. When a serotype was not presented,
it was unclear whether it was:
Some countries with large under-five populations had limited or no data in the analysis, but serotype data was often avail-
able from neighboring countries. A large number of isolates collected over an extended period of time is required to have
robust country level estimates, and in the absence of additional studies with serotype data that meet these characteristics,
it is unlikely that additional serotype data will change the region level inferences from this analysis, although it may affect
sub-regional inferences.
”Antiretrovirals are only part of the solution despite their ability to cause a
tremendous decline in invasive pneumococcal disease. Even with access to
antiretrovirals, HIV-infected individuals in the United States are still at 35-fold greater
risk of developing invasive pneumococcal disease than the general population.”
— SHABIR MADHI , U n iv er s it y o f W it wat er s r a n d, S o u t h A f r i c a
In sub-Saharan African countries where the burden of HIV is greatest, the focus of research in terms of prevention of pneu-
mococcal disease in HIV-infected adults has centered on the use of the pneumococcal polysaccharide vaccine. In the United
States, a longitudinal cohort study of 24,000 individuals showed about a 20 percent reduction in pneumonia, as well as
invasive pneumococcal disease among HIV-infected adults who received the pneumococcal polysaccharide vaccine over the
past five years compared to those who didn’t. This study also showed that in the months following vaccination with the pneu-
mococcal polysaccharide vaccine, a heightened susceptibility to pneumonia and invasive pneumococcal disease occurred in
the vaccinated group.
The data from the CDC probably gives some explanation for the differences that are observed in terms of the efficacy studies
in the United States compared to Gambia and Uganda. In the United States, researchers found that vaccine effectiveness is
greatest irrespective of antiretroviral status amongst those individuals in whom the viral load is less than 100,000. In indi-
viduals with a viral load of greater than 100,000 there is a trend toward increased risk of developing pneumococcal disease,
except for those with a CD4 count of greater than 500.
Data showing the impact of antiretrovirals on opportunistic infections and bacterial infections in HIV-infected children is
sparse, and this is the one study which looks at the incidence of opportunistic infections compared to historical control.
The study from the United States shows a decrease in bacterial pneumonia as well as a decrease in bacterial bacteremia.
However, this analysis is problematic because it does not adjust for changes in terms of co-trimoxazole prophylaxis or the
introduction of pneumococcal conjugate vaccine.
The pneumococcal conjugate vaccine has been evaluated in HIV-infected children across several settings. In a study from the
United States where 80 percent of the children received antiretrovirals during the time of primary vaccination, results showed
that after the primary series of vaccination, and both before and after the booster dose of pneumococcal conjugate vaccine,
the quantitative antibody responses to the pneumococcal vaccine was exactly the same as for children not infected with HIV.
In a similar setting in South Africa, although children were not receiving antiretrovirals, there was a trend toward lower
quantity of antibody, but this was only significant for serotypes 1 and 18C. Although not statistically significant, the rest of
the serotypes showed a trend toward lower GMCs compared to the children who were not infected.
Although the efficacy of the vaccine is less in HIV-infected children, it is still important to note that burden of disease
prevented in HIV-infected children is 18-fold greater than for non-infected children. Also, in pneumonia, although the vaccine
efficacy is less against pneumococcal pneumonia in HIV-infected children, the absolute burden of disease being prevented in
HIV-infected children for pneumococcal pneumonia is at least nine-fold greater.
Five years after vaccination, HIV-infected children have lower antibody concentrations compared to kids not infected by HIV.
2302
1000
500 267
Unfortunately, there is a falloff in terms of efficacy from 65 percent for those vaccine serotypes after two years of follow-up,
down to 39 percent after five years of follow-up. In the absence of a booster dose of the vaccine, there is some loss of protec-
tion, which is now raising questions within South Africa in terms of the dosing schedule.
The pneumococcal conjugate vaccine has also been studied in HIV-infected adults, and four different studies show a fairly
positive finding in terms of immunogenicity of the vaccine and quantitative and qualitative responses, and no impact in
terms of change in viral load with these T-cell dependent immune responses.
Early preliminary data presented by Neil French shows a very positive outcome, which is that in HIV-infected adults that
previously had invasive pneumococcal disease randomized to the vaccine and a placebo, there was a 75 percent reduction in
vaccine serotype invasive pneumococcal disease. However, that is tempered to the diversity of serotypes that cause invasive
pneumococcal disease in HIV-infected adults, and the net effect is a 28 percent reduction in invasive pneumococcal disease.
The introduction of a pneumococcal conjugate vaccine for HIV-infected adults will probably not take place until five or 10
years from now. The likelihood of the vaccine reducing invasive pneumococcal disease in HIV-infected adults relates to the
potential indirect protection of the vaccine in unimmunized communities. In the United States, data on indirect protection
shows a 65 percent reduction following the introduction of pneumococcal vaccination of young children.
For adults in the United States, a 62 percent reduction in invasive pneumococcal disease amongst HIV-infected adults is due to
the vaccine serotypes. Unfortunately, a 44 percent increase in invasive pneumococcal disease is due to the non-vaccine sero-
types in HIV-infected individuals, something which is not seen to the same extent amongst the non-HIV infected population.
The net effect was a 20 percent reduction in all invasive pneumococcal disease, resulting in an absolute burden of disease
that has been lowered to 100–800 cases per 100,000. In countries where antiretrovirals are not in use but there is indirect
protection, you would expect this absolute burden of disease that has been prevented, even with a 20 percent reduction, to
be much greater.
A recent publication by Vanna Albrecht shows a 40 percent reduction in invasive pneumococcal disease and invasive pneu-
monia among HIV-infected adults. However, in HIV-infected adults with full-blown AIDS, there is an increase in invasive
In conclusion, the pneumococcal conjugate vaccine undoubtedly has a role to play in reducing the burden of invasive pneumo-
coccal disease in children. However, the degree to which protection is sustained in these individuals needs to be addressed.
The introduction of antiretrovirals could change the functionality and the durability of antibody responses of the conjugate
vaccine, and it is research that is being actively pursued. This is something which has been scaled up in South Africa over
the course of the past two to three years. However, despite the national program of antiretrovirals, less than 20 percent of
HIV-infected children in South Africa who acquire antiretrovirals actually receive them.
It is a known fact that about 170,000 children die of meningitis every year, and that two-thirds of these deaths are in devel-
oping countries. In more developed places, nearly 20 percent of children get some sequelae from meningitis. Were noted that
suspicion that this number is much higher in the developing world and in Africa led the researchers to undertake this study.
Methodology
The literature review for bacterial meningitis sequelae in the African region selected studies which were cohorts, cross-
sectional, case control, intervention, or clinical trials, and reviewed articles with original data. Specific information and data
for children between one month and 15 years and the etiology of meningitis had to have been determined by culture or rapid
antigen testing to meet the review criteria.
Six articles were selected, representing 6,230 patients across Africa. From these articles the researchers sought early
in-hospital results, including death, and the sequelae that were identified. A few studies involved a follow-up that was
comprehensive enough for the researchers to conduct a post-discharge evaluation of about 1,240 of the patients.
Mortality results from these studies show that in-hospital deaths for S. pneumoniae have a median case fatality of 35 percent,
which is above the median for all meningitis. From the local evaluation, S. pneumoniae is the most lethal form of pneumonia
within the immediate period of care.
From the few post-discharge studies, mortality was shown to be about 13 percent, a rate higher than the other three common
causes of meningitis. More than a quarter of patients available for follow-up showed signs of cerebral palsy, one-fifth had
cognitive defects or mental retardation, nearly 40 percent had some hearing loss, and 14 percent showed some visual loss.
Mortality 13 11 0
Physical Defects 26 12 1
Cognitive Defects 20 14 0
Hearing Loss 37 21 2
Visual Loss 14 2 0
In conclusion, within the limitations of the available studies, S. pneumoniae and meningitis appear to burden Africa with very
high mortality and neurological sequelae. The societal cost of caring for surviving children with sequelae is enormous and
unlikely to be optimally met by financially constrained populations. Furthermore, such children are also much less likely to
live to adulthood.
The researchers investigated the risk of pneumonia, meningitis, and bacteremia in Africans with sickle cell disease compared
to those without sickle cell. Africa bears the greatest burden of hemoglobin disorders, with almost a 250,000 infants born
each year with sickle cell anemia. That accounts for almost two-thirds of all total global births affected by severe hemoglobin
disorders annually.
On average, it is estimated that 18 percent of the African population carries a significant hemoglobin variant such as hemo-
globin S, C, or beta thalassemia. The countries in Africa most affected are in the Equatorial region where malaria is endemic.
These populations have the highest rate of sickle cell anemia, and about 1–2 percent of all births there are homozygous for
hemoglobin S.
Global distribution of
haemoglobin disorders, in
terms of births of affected
infants per 1000 births
(WHO, 1995)
< 10 5 - 9.9
0.1 - 0.19 10 - 18.9
0.20 - 0.99 < 19
1 - 4.9
The outcomes of interest that were examined were meningitis with a suspected, presumed, or confirmed bacterial etiology;
pneumonia with a suspected, presumed, or confirmed bacterial etiology; and bacteremia, which was defined as non-pneu-
monia, non-meningitis, or NPNM disease with a positive blood culture plus or minus another focus of infection.
A systematic literature search was conducted using eight databases and keywords relating to sickle cell disease, hemoglo-
binopathy, febrile infection, bacteremia, pneumonia, and Streptococcus pneumoniae. In addition, literature databases were
hand-searched, and the results from the meningitis review that Dr. Were presented were hand-screened.
This resulted in 272 unique citations screened by preset inclusion and exclusion criteria. These citations were then narrowed
to 32 articles from 12 different countries.
After finding a wide variability in study design and methodology, the researchers conducted a formal quality scoring of the
studies. This provided a way to compare studies, as well as identify as a group any major quality concerns or data gaps that
these studies presented.
Data Limitations
Some of the data gaps identified affected how the researchers were able to interpret results and eventually how they deter-
mined which studies addressed the primary research question. One area of data concern was the sparse data available for
There was also very little data available on prior antibiotic use among study subjects which, in particular, may be a common
occurrence in sickle cell patients with their frequent infectious complications. Only four articles addressed this issue of
pre-hospital antibiotic use. In those four studies, the prevalence of pre-hospital antibiotic use ranged from 15–48 percent
of subjects.
The biggest quality concern was that most studies did not have a control group, either of non-sickle cell patients or patients
without bacterial disease. Although the researchers found many articles on the bacterial etiology of disease, they were
unsure how to interpret the results without a control group, and thus unable to calculate a relative risk in most cases.
The majority of identified articles had methodological limitations that made it difficult to assess risk between sickle cell
disease status and invasive bacterial disease. The main limitation was the lack of a control group, which is more important
because labs may not pick up fastidious organisms such as pneumococcus even when it is present. Within the studies that
were designed to assess risk association, the data is consistent and showed increased odds of sickle cell disease genotypes
among cases of disease compared to the prevalence of sickle cell disease and the reference or control populations.
There are some additional limitations and potential biases worth noting. There is a selection bias, as most of the studies
used were conducted in teaching hospitals or tertiary hospitals in urban settings, so there was a bias toward patients with
relatively better access to care, who do not die early from their infectious complication, and in some cases, who were able to
pay for the diagnostic tests and get treatment.
There is also the potential for ascertainment bias, as sickle cell disease testing may not have been done in a random manner
among all of the bacterial cases, but rather in some purposive way, when clinicians have a prior suspicion that a patient had
sickle cell anemia.
Results
The articles found through the literature search were varied in study design and methodology, so a standardized quality
scoring form was designed based on six criteria:
1) the relevance of the article’s study question(s) to the objectives of this review;
2) rigorous and clearly described participant selection methods;
3) use of a standard approach for ascertainment of sickle cell disease (SCD) and ascertainment of bacterial infection
among participants;
4) inclusion of a control group (subjects without bacterial infection or without SCD);
5) exploration of the statistical significance of study results, if appropriate; and
6) overall impression of study.
There were six articles that met the quality criteria and were designed in a way that explored the association between hemo-
globin status and the risk of invasive bacterial disease. All six of these articles had a control group or a reference population.
Four of these studies were from the Democratic Republic of the Congo or former Zaire, one study was from Nigeria, and one
from Senegal.
The studies were consistent in the results that they found, namely, that there was an increased risk association between
sickle cell disease and invasive bacterial infection. The pooled odds ratio (OR) for sickle cell disease among meningitis, pneu-
monia, and NPNM patients was much higher than the sickle cell disease prevalence in the reference or control populations,
and the highest association was seen between sickle cell disease and invasive pneumococcal disease.
For sickle cell disease among meningitis, pneumonia, and NPNM patients, OR was higher than for the reference population, and
the highest association was seen with invasive pneumococcal disease. For all cause laboratory-confirmed bacterial disease,
the studies had individual ORs which ranged from nine to 22, but the pooled OR was 17. For invasive pneumococcal disease,
the pooled OR was higher at 36.5, and for invasive Haemophilus influenzae type B or Hib disease, the pooled OR was 8.7.
A few studies focused specifically on meningitis subjects and the syndromes were examined separately. Two of the studies
combined all-cause laboratory confirmed bacterial meningitis and yielded an OR of 20. The pooled OR for pneumococcal
meningitis was a bit higher at 28, and the OR for Hib meningitis was 9.2. All of these ORs were statistically significant.
In conclusion, Africa bears the greatest burden of pneumococcal disease with sub-Saharan Africa accounting for almost half
of all childhood pneumonia deaths, while, in contrast, only comprising one-fifth of the world’s under five population. The
same region accounts for almost two-thirds of all births affected by severe hemoglobin disorders, so this combination of
factors presents a major public health concern for most African countries.
In contrast, there is not as much clinical and preventive support for sickle cell specific services in Africa. A recent report from
WHO and the Thalassemia International Federation found that in most of Africa, outside of major urban hospitals, clinical and
screening services are severely limited. There is also limited health information available to the general public, while treat-
ment costs are borne by families and — in the context of a genetic disorder — put a dire burden on affected families.
Looking ahead, priorities will be to improve duration and quality of life for African children with sickle cell disease. An
emphasis on prevention of infectious complications would help close the quality gap between sickle cell disease services
available in high-income countries and most African countries.
Discussion
Orin Levine led a wide ranging discussion in which members of the audience asked the researchers to elaborate on a number
of topics.
QUESTION: Bill Hausdorff of GlaxoSmithKline asked if the incidence of serotype 1 in diseases in older children was truly
higher than those in younger children. He speculated that it’s proportionally higher in older kids because the other serotypes
are disappearing by that age.
RESPONSE: Meena Ramakrishnan responded that the type of analysis that Hausdorff was suggesting would require raw
datasets, but that it is an analysis the research team would like to conduct in the future. Ramakrishnan also noted that the
researchers suspected an interaction between age and syndrome, although they did not report these results in the study
RESPONSE: Ron Dagan from Israel added that when an analysis of serotypes is conducted, researchers typically organize data
by age or by continent, but in effect, when the data is broken down by entities and by type of diagnosis, differences arise. He
notes that it is commonly accepted that serotype 1 is not an infant disease. Also, it is generally accepted that pleural pneu-
monia in older kids and young adults is associated more with serotypes that are not commonly found in very young infants.
Taking all of this into consideration, Dagan suggested that at least some of the sub-analysis should be conducted in the
manner in which Ramakrishnan did by identifying specific entities and conducting a sub-analysis of serotypes. Currently,
there is not enough data to say which type of entities can be prevented with which vaccine. This will hold true especially in
the future when there will be different vaccines formulated for different serotypes.
QUESTION: Richard Adegbola with Medical Research Council Gambia directed a question to Fred Were concerning the
mortality rate reported for Neisseria meningitidis. Adegbola noted in his question that it comes in epidemics, particularly
where he works and that during these epidemics, mortality risk can actually be as high as 20 percent. Knowing this, does he
think the 4 percent mortality rate that he reported might be giving a misleading indication to people that it’s not an impor-
tant disease?
RESPONSE: Fred Were responded that although it seems misleading, there was not enough data to provide a better answer.
The only articles that reported mortality for meningitis were from North Africa, and there, the outcome is probably in the
beltway of many of these epidemics.
QUESTION: Peter Paradiso from Wyeth directed a question to Shabir Madhi regarding the disease patterns that relate to
vaccine serotype versus new emergent serotypes in HIV-positive children and adults. Paradiso suggested that some of the
patterns that Madhi showed in the United States are a combination, but there are more new serotypes not in the vaccine,
particularly 19A. He asked Madhi if he has seen emerging serotypes in his population since the trial and over the last five years.
Secondly, he asked him to comment on the importance and practicality of a booster dose in these populations in his country.
RESPONSE: Madhi said that in terms of serotype changes, the large analysis was conducted around 2005, and what was
notable was that there was not an emergence of new serotypes in HIV-infected adults. Pediatric serotypes, the serotypes
included in PCV7, are much more dominant in HIV-infected adults, especially amongst mothers. This is not as surprising in
South Africa because 70 percent of households are single family households, so the contact between mother and children
is much stronger than the contact between father and child. This explains the dominance of the traditionally pediatric sero-
types in HIV-infected individuals. There hasn’t been much change in terms of, as an example, serotype 19A. This is going back
to data from 2005, although there may be more information available based on more recent data.
In terms of what is being done for a booster dose of the vaccine, Madhi noted that the one thing that he omitted from the
presentation was that five years after the primary series, HIV-infected children don’t mount a better response to a boosted
dose of the vaccine compared to HIV-infected children who didn’t receive a vaccine during infancy. For that reason in South
Africa, a program of 2+1 with a boosted dose to be provided at nine months of age has been implemented. So it will be a six
week and 14 week schedule together with a boosted dose at nine months, which should extend protection, but won’t address
the issue as to how long that protection is going to last.
QUESTION: Dan Efigan from CDC in Kenya asked a question directed to Shabir Madhi about the indirect effect on HIV-positive
adults. He noted that in the United States, that impact seemed to be greater than among the general population and asked
Madhi if this was due to serotype distribution or age distribution and exposure to children. He also asked Madhi to speculate
on what he thinks the indirect impact in that population of HIV-positive adults would be if this was brought to Africa.
RESPONSE: Madhi noted that it’s nearly impossible to extrapolate from the findings of the United States to Africa for a
number of reasons. The data from the United States which he didn’t show was that the net effect of reduction in invasive
disease in terms of indirect protection was least amongst women, and the hypothesis is that it is because the epidemiology
of HIV infection in that group of individuals is probably related more to sexual transmission rather than to intravenous abuse
or other factors.
Madhi noted that in South Africa the epidemic is driven primarily by sexual transmission of HIV, and that particular group of
individuals in the U.S. probably reflects more closely what might happen in South Africa. However, the big issue of indirect
protection in countries in Africa is that the direction of transmission of pneumococcus is not as clearly established as in
the U.S.
In countries like Gambia, for example, 50 percent of adults who are not infected are actually colonized with pneumococcus
compared to only about 10–15 percent in the United States. Madhi noted that he is currently conducting colonization studies
in South Africa that show a very high prevalence of colonization with those vaccine serotypes, especially in women.
The prevalence of colonization in women is much higher, especially where a catch-up campaign is not in place in South Africa.
The effects of indirect protection will not be realized for awhile, and that indirect protection will be completely based on
assumption that the direction of transmission is from infant to adult rather than bidirectional. Dr. Madhi said “This doesn’t
give you a direct answer, but I really think there is a big, big question mark in terms of indirect protection.”
“I think another big issue is that the problem with the population pyramid in Africa is very different from the United States,”
Madhi said. There is a higher burden of disease, but the population pyramid is very different so there are few individuals 65
years of age or older in Africa. For this reason, the type of indirect protection that you see in terms of absolute numbers in
the United States might not be what would be realized in other settings such as Africa.
RESPONSE: An unidentified speaker added that in addition to the issue of HIV and directional flow, indirect effect and protec-
tion need to be considered in terms of transmission. For example, in order to be really protected, if you give a 7-valent vaccine
and reduce and replace the vaccine serotypes in the nasopharynx with non-vaccine serotypes, the non-immunocompromised
will be protected because they are less virulent. The more immunocompromised will be less protected because the least
virulent bugs can still cause disease. If you are going to see the replacement in the nasopharynx, then the HIV population
might be less protected indirectly than the rest of the population, which has to be added as a factor, too.
RESPONSE: Madhi added that he was in full agreement and also pointed out that in the study that involved immunocompro-
mised HIV-infected individuals, there was no indirect protection and there was an increase in disease. The fact that these
pediatric serotypes cause some disease in non-infected adults in South Africa but much more disease in HIV-infected adults
also shows an increased susceptibility to these supposedly virulent serotypes in HIV-infected individuals.
RESPONSE: Meena Ramakrishnan commented that it would be great to consider the population of sickle cell patients ahead
of time and to use any further vaccine surveillance studies done in Africa, as well as any other vaccine efficacy trials, to help
strengthen the case. The limitations of surveillance and laboratory detection of pneumococcus were mentioned earlier, and
it would be important to consider sickle cell patients, like HIV patients, as a special population within a vaccine trial. Even
without that specific data, Ramakrishnan added, “I think we do have good data from not only high-income countries but
countries from the Caribbean that do show the efficacy of vaccinating sickle cell patients.”
In terms of the schedule, Ramakrishnan suggested that it doesn’t warrant a separate consideration apart from the EPI
schedule and what would best fit into the delivery of health services. In general, this would be another way of supporting the
clinical care and education and prevention of complications in sickle cell patients where there is a huge burden of disease
and a severe lack of services. Targeting some preventive approaches would help close some of the quality gap between
services available to high-income sickle cell patients or those in the Caribbean, where life expectancy is 40–50 years of age.
The reality for most African populations is that there is little data on life expectancy, but it’s generally thought to be less
than 20 years of age in sickle cell subjects.
RESPONSE: Madhi added that this question is of particular importance, because as more antiretrovirals are being rolled out,
HIV-infected children will survive for much longer periods of time. Also, as the data showed, at least in the United States,
the background rate of invasive pneumococcal disease in these children is going to be less than the general population but
still a large increase. Madhi added that the strategy in terms of vaccination implementation does not warrant a catch-up
campaign. In fact, resource constraints necessitate vaccinating children at six and 14 weeks, and these are the children who
will be getting the vaccine at nine months of age. There is no plan for the catch-up campaign anytime soon, in addition to
which there is no plan for targeted catch-up campaigns within HIV-infected children.
Madhi presumed that the net effect will be that the type of dramatic decline that was seen in invasive pneumococcal disease
in the United States is not going to materialize in South Africa for at least three to four years. He attributes this to the fact
that there is a large pool of HIV-infected children who will remain unvaccinated, and therefore remain at risk of invasive
disease. In terms of planning, resource constraints have been a major limitation in optimizing the immediate benefit that
could be gained by vaccinating HIV-infected children.
With the fourth Millennium Development Goal (MDG) calling to reduce child mortality by 2015, continued advancement of pneu-
mococcal conjugate vaccines has become a global priority. The production of the 7-valent pneumococcal conjugate vaccine
(PCV7) for global distribution is an extremely important step to achieving this goal. Three presentations addressed the
impact of vaccines in the Republic of the Gambia, South Africa and the United States. They highlighted overall efficacy against
pneumococcal disease, but also underscored the need for continued research and progress to develop a vaccine that would
eliminate pneumococcal disease serotypes and replacement serotypes. A fourth presentation examined data pertaining to
serotype replacement and provided a call for continued progress toward the elimination of pneumococcal disease.
In the first presentation, Dr. Richard Adegbola, Head of the Bacterial Diseases Programmed for the Medical Research Council
(UK), shared the Gambia’s experience with the pneumococcal conjugate vaccine. While vaccination against certain serotypes
in the Gambia has not matched the efficacy and success of other global trials, he emphasized that use of the vaccine in the
Gambia did relieve the overall burden of pneumococcal disease in children and in adults. PCV7 has a 50 percent efficacy in
the Gambia. Although he is pleased that an immunization program with PCV7 will begin in the Gambia this summer, he called
for continued research and vaccine testing.
Beyond the Gambia, efficacy trials of pneumococcal conjugate vaccines were successful. Keith Klugman, Chair of Global
Health at Emory University, discussed the efficacy of the nine-valent pneumococcal conjugate vaccine (PCV9) in South Africa
and also provided evidence that pneumococcal conjugate vaccines prevent other invasive diseases. Specifically, he discussed
the relationship between the pneumococcal conjugate vaccine and influenza. Klugman explained that beyond preventing
pneumococcal disease, the vaccination could help prevent an outbreak of pandemic influenza.
Finally, Cynthia Whitney of the U.S. Centers for Disease Control and Prevention (CDC) praised the success of PCV7 in the United
States, which has reduced mortality due to pneumococcal disease among both children and adults. She also noted that the
benefit has not come only from the immunizations, but also as a result of herd immunity. Unfortunately, a remote region in
Alaska suffers from high rates of pneumococcal disease, despite the vaccination. This is a case of replacement with serotype
19A. Whitney called on her peers to come together to develop a response to the problem of replacement with serotype 19A.
Ron Dagan, Professor of Pediatrics and Infectious Diseases at Ben-Gurion University of Negev in Beer-Sheva, Israel, continued
the conversation around serotype replacement, emphasizing that rates of pneumococcal disease remain low, despite sero-
type replacement. He argued that serotype replacement mostly occurs with weaker, less virulent serotypes contributing to
lower rates of the overall pneumococcal disease. One important exception to this happens with replacement by serotype 19A,
which Whitney showed affects Alaska. Serotype 19A is multi-drug and penicillin resistant, and when it occurs, it can cause the
overall rate of infection to increase or remain unchanged by vaccination.
Adegbola shared that some of the diseases studied by the Greenwood team are no longer a primary concern in the preven-
tion of childhood death in the Gambia. The Gambia is fortunate to have benefitted from various immunization programs
throughout the 1980s and 1990s. Specifically, coverage by the Expanded Program on Immunization (EPI) in the Gambia is one
of the highest in all of Africa, with the vaccines causing a decrease in approximately 82 percent of the target antigens. Addi-
tionally, implementation of a vaccination program in the Gambia for Hib has almost entirely wiped out the disease. Overall,
there has been a dramatic decrease in life-threatening disease among children in the Gambia. Yet, pneumococcal disease
remains a potent threat.
A series of community-based studies was launched throughout the Gambia seeking to identify the immunogenicity and
safety of various pneumococcal conjugate vaccines. With these studies showing positive results — the vaccines were safe,
they induced an immune response and they did not interfere with EPI vaccines — a randomized, controlled efficacy trial of
PCV9 was launched. Between 2001 and 2004, PCV9 was provided to 17,437 children at two, three and four months of age. The
health of the immunized children was monitored for the 30 months following the final immunization.
The study established that PCV9 was safe, effective and didn’t interfere with other disease vaccinations in the Gambia. Specif-
ically, the vaccine showed 37 percent efficacy against radiological pneumonia and, more importantly, it showed a 16 percent
reduction in overall mortality, a 15 percent reduction of overall hospital admissions and a 7 percent reduction in clinical
pneumonia. Adegbola praised PCV9 for displaying approximately 80 percent effectiveness in rural Africa, where access to
treatment for childhood illness is low. The study was the first in 20 years to show such an impact on infant mortality. With
such results, Adegbola explained that “the preventable burden of the disease seems higher” and discussed his hope that
herd immunity would boost the vaccine’s effectiveness, as it has elsewhere in the world.
The Pneumococcal Surveillance Project in the Gambia is preparing to evaluate the effectiveness of PCV7. The goal of this
study is to calculate the incidence of the disease, collect and monitor different serotypes of the disease and monitor serotype
replacement, if any occurs. Eventually, this data will be aggregated for before and after comparisons of what is happening
with pneumococcal disease in the Gambia.
Beyond the promise of immunogenicity of PCV9, Klugman’s studies uncovered another important aspect of the vaccine’s
effectiveness. Klugman explained one study during which a group of children achieved the necessary protective levels of
immunization against the disease after receiving only two doses of the vaccine. This knowledge provided confidence that a
two dose primary vaccination schedule will protect children against the disease until they receive the nine month booster.
PCV9 also proved to be effective in preventing invasive pneumococcal diseases in both immunocompromised and in rela-
tively healthy children. Klugman described the positive results of a study in Soweto, during which PCV9 was 85 percent
effective in protecting invasive pneumococcal disease in children not infected by HIV. Yet, in children who were HIV-positive
and therefore immunocompromised, the vaccination was 60 percent effective. Although the vaccine was less effective for
children with weakened immune systems, this evidence translated into significant disease prevention.
Antibiotic resistance is a concern when developing immunizations of all diseases. Klugman explained that continued exposure
to PCV9 led to antibiotic resistance in non-vaccine serotypes. In a trial study, Klugman showed that the impact on antibiotic-
resistant invasive disease was approximately 60 percent. The carriers of the resistant disease were typically HIV-infected.
The relationship between pneumococcal disease and influenza is extremely important in the sphere of public health. Klugman
shared his findings that data from the 1918 flu pandemic indicates that most of the deaths were due to pneumococcal-
associated influenza. Pandemic influenza remains one of the largest threats to mankind. With data supporting this relation-
ship, Klugman argued that continued advancement of the vaccine could prevent widespread mortality due to an epidemic
similar to the 1918 influenza outbreak.
With all of the promise studies provided that pneumococcal conjugate vaccines are immunogenic and safe in South Africa,
Klugman called upon his peers to continue their efforts toward the improvement of the vaccine. Similar to the Gambia, PCV7,
which is being produced, is much less effective within South African populations than PCV9. Improvements in the vaccine for
South Africa could mean not only lowered levels of pneumococcal disease, but lowered levels of other invasive disease in the
region and could be critical to the prevention of a pandemic invasive disease outbreak.
Figure 8 | Rates of Invasive Pneumococcal Disease Among Children <5 Years, 1998–2007
120
Overall
100
Cases per 100,000
PCV7
40
Introduced 22-25 cases per 100,000
20
0
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Year
The vaccine also proved to be effective for children with weakened immune systems. Children with sickle cell anemia who
received the vaccine also benefited, with a much lower occurrence of invasive pneumococcal disease. This study provided a rate
of vaccine effectiveness in children with sickle cell anemia of more than 81 percent, after one or more doses of the vaccine.
Serotype Replacement
Although Whitney celebrated the effectiveness of PCV7, she emphasized that the results do not translate into elimination of
all serotypes by vaccination. While this vaccine is effective against approximately 80 percent of invasive disease in the U.S.,
studies indicated that there was also approximately a 37 percent increase in non-vaccine serotypes.
Although replacement exists in the United States, it does not affect the majority of the population. For the children of
Alaska natives in a particular remote region, however, replacement is a real problem. Whitney explained that pneumococcal
serotypes PCV7 disappeared after vaccination, in place of these serotypes; serotype 19A has emerged at an increasing rate.
Elsewhere in the United States, children and adults have experienced some replacement. Whitney is unable to explain why
the children from this remote Alaska region are afflicted with this replacement disease.
Whitney points out that replacement by serotype 19A should be closely monitored. The emergence of serotype 19A has
increased almost 100 percent since immunization campaigns began in the United States eight years ago. Whitney challenged
her peers to continue studying the vaccine and why some non-vaccine serotypes, such as 19A, are more problematic in some
populations than in others.
70
65+ years
60 PCV7
Cases/100,000 population
2007 vs.
50 baseline
40 -37%
30
40-64 years
20
18-39 years -23%
10
-49%
0
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Figure 10 | Invasive Pneumococcal Disease Rates Alaska Native Children < 2 Years Old
350
PCV7 type
non-PCV7 type
300
Rates per 100,000
250
150
100
50
0
1995-2000 2001 2002 2003 2004 2005 2006 2007
Replacement occurs when certain serotypes of the disease disappear and other, new serotypes develop.
Dagan discussed that replacement likely occurs in the nasopharynx. Nasopharyngeal replacement could have one of
three outcomes:
1. The new disease could be more powerful,
2. The new disease could be less powerful, or
3. The new disesase could have similar effects as the disease it replaced.
Dagan showed evidence that the nasopharyngeal replacement serotypes are less virulent, which results in a sustained
decline in the rate of disease infection. While replacement threatens the development of a stronger, non-vaccine
pneumococcal serotype, the overall effect of replacement has been a sustained reduction in the rate of the disease.
90
PCV7 type
80
non-PCV7 type
Cases/100,000 population
70
60
50
PCV7 Introduced
40
30
20
10
0
1998 1999 2000 2001 2002 2003 2004
Year
Hick et al, J Infect Dis, 196:1346-54, 2007
While pneumococcal disease is most often transmitted by children and has a higher mortality rate in children, it is also very
important to understand what replacement of serotypes occurs in adults. He argues that rates of serotype replacement are
higher in individuals who are immunocompromised, such as the elderly or people who are HIV-positive. There is no evidence
to suggest that replacement occurs in HIV-negative adults between 18 and 64 years of age, likely because these individuals
are strong and healthy.
Figure 12 | % Pnc from NP of Healthy Children that Were VT, Vaccine-Related or Non-VT, By Quarter
(2000–2003)
25%
% Pnc
20%
15%
10%
5%
0%
1 2 3 4 5 6 7 8 9 10 11 12
Quarter (2000–2003)
Pelton et al, Pediatr Infect Dis J, 23:1015-22, 2004
The development and introduction of PCV has not caused the rate of infection by 19A to increase in frequency, nor did it cause
the serotype to become multi-drug and penicillin resistant. Dagan used examples from Korea and Spain where the frequency
and severity of serotype 19A increased before vaccinations were introduced. He also used examples of western countries
where the rates of infection by serotype 19A increased rapidly in both vaccinated and non-vaccinated areas. Dagan therefore
stressed the importance of continued study on serotype 19A so that a new conjugate vaccine could be developed that would
protect against it.
Progress toward preventing pneumococcal-related child mortality continues. Dagan proved that replacement disease should
not be ignored and that it poses a threat to vaccinated populations. “Evidence for nasopharyngeal replacement after PCV is
universal and beyond any doubt,” he says. But he also agrees that replacement has had limited impact on overall rates of the
disease and other invasive pneumococcus-related disease.
Discussion
QUESTION: William Hausdorff from GlaxoSmithKline Biologicals noted how the efficacy studies done in the Gambia and South
Africa did not include a booster after the first three immunizations, while studies done in the U.S. did include a booster at
twelve months. He continued to ask if a booster in the Gambia might increase the efficacy of PCV9, specifically if it would
improve the ability to protect against serotype 1.
ANSWER: Keith Klugman responded to Hausdorff’s question. A boost for vaccinations in the Gambia was not considered.
Klugman expressed his optimism, however, that a conjugate vaccine would be developed to protect against serotype 1.
Klugman indicated that research he has conducted on serum therapy in the early 1920s shows convincing evidence that effi-
cacy against serotype 1 is possible. More important, Klugman said, is the question of whether the immunization of children
will prevent the overall transmission of serotype 1.
QUESTION: Peter Paradiso from Wyeth Vaccines asked if they did functional antibody responses.
ANSWER: Klugman responded that functional antibody responses were completed, but not to serotype 1 in the Gambia studies.
Paradiso continued by asking if there were functional antibody responses to the other serotypes. Klugman responded that
there were responses, but that they have not been directly compared.
The fourth session featured four presentations that discussed the development of next-generation pneumococcal vaccines
and the challenges to development. Peter Paradiso from Wyeth Vaccines and William Hausdorff from GlaxoSmithKline Biologi-
cals discussed specific vaccines that are approaching licensure. Mark Alderson, from the Pneumococcal Vaccine Project
at PATH, followed with a discussion on the steps that PATH is taking to support the development of other next-genera-
tion pneumococcal vaccines through several partnerships. On behalf of the Developing Countries Vaccine Manufacturer’s
Network, Akira Homma of Bio-Manguinhos/Fiocruz addressed the challenges associated with the production of pneumo-
coccal vaccines, and their effects. Specifically, Homma noted the high cost of production and the high cost of distribution in
middle-income countries.
Wyeth’s PCV13 builds on the protection provided by the successful PCV7. In addition to protecting against all seven serotypes
targeted in PCV7, it produces an immunogenic response against six additional pneumococcal serotypes, including one, three,
five, six A, seven F and 19A. Overall, studies show that a regular immunization schedule with PCV13 invokes a high antibody
response that is comparable to PCV7’s rates of effectiveness. While PCV13 immunogenicity results were sometimes slightly
With tests showing no clinically significant differences between PCV7 and PCV13 in the overall spontaneous adverse effects,
Paradiso applauded the vaccine’s safety. Beyond its safety, Paradiso also shared the advantages of the vaccine — notably
that transitioning from PCV7 to PCV13 is easy and efficacious. Because all of the serotypes represented in PCV7 are also
represented in PCV13, it is possible to substitute PCV13 for PCV7 at any point in the vaccination schedule, without negatively
affecting the immune response.
Paradiso said he expected to receive licensures of PCV13 by the end of 2009, with applications having been submitted in
Europe, South Africa, Canada and elsewhere. In planning for future PCV, Paradiso acknowledged his belief that Wyeth has
reached the limit in terms of the number of serotypes added to one pneumococcal vaccine. Future next-generation conjugate
vaccine research will likely need to consider different formulations of PCV13.
45
40
5
1
% of all IPD each year
35
30
25
20
15
10
0
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Before closing, Hausdorff reiterated two important advantages of a vaccine with a carrier protein from a different pathogen.
First, Hausdorff explained that it minimizes the risk of immune interference with co-administered vaccines. Furthermore,
he continued by emphasizing the importance of the additional protection the vaccine would provide against disease by the
second pathogen.
To obtain licensure through WHO and European Regulatory Authorities, GSK used antibody comparisons with a licensed
vaccine, in this case PCV7. PHiD-CV has been licensed in one country and has received a positive opinion from the European
Regulatory Agency. This is the first global 10-serotype pneumococcal conjugate vaccine with a Haemophilus protein and is
the first pneumococcal vaccine to meet the WHO target profile.
The main goal of PVP is to accelerate the development of safe, effective and affordable vaccines against Streptococcus
pneumoniae for children in Africa and other low-income areas of the world. Focusing its work through six main objectives and
working with various partners to accomplish those objectives, PATH supports numerous research initiatives.
Protein Conjugate
subunit vaccine
(Mucosis) (Partner 2)
Protein Killed
subunit whole cell
(University (CHB/Butant
of Adelaide) an Institute)
= PATH-supported
Protein
subunit = not PATH-supported
(University = under PATH consideration
of Glasgow)
Homma discussed the inherent challenges to the incorporation of PCV7 in national immunization programs in middle-income
countries. Mainly, the vaccine is too costly. Without the production of additional, safe and effective pneumococcal vaccines,
middle-income countries may be unable to afford it.
Homma explained how costs affect high, middle and low-income countries differently. The cost of the vaccine is different
for countries depending upon their economic status. Data shows that high-income countries are unaffected by higher-cost
vaccines between $55 to $60, and lower-income countries are also unaffected by cost of the vaccine, as it has been lowered
to $7 per vaccine. In middle-income countries, however, costs remain at $20 per vaccine, which continues to be prohibitive
for many.
To lower cost, there must be competition between vaccines. Although several groups presented next-generation vaccines in
development, it will take a long time for these vaccines to be produced. Unfortunately, Homma also discussed that the cost
of production is also very high and creates another barrier to the production of vaccines. Conjugate vaccine plants are very
expensive and consist of complex technology and complex regulatory requirements. Homma said that data from three to four
years ago provides evidence that “a new plan of production will require an investment of up to $180 million in construction,
building and organizing.” Homma shared that the DCVMN is working to strengthen relationships and partnerships within the
research industries that could help to cut the cost of new technology development.
Price/Dose
$60 $20 (?) $7 (?) —
(average est.)
Vaccine Market
40 130 135 305
(Mi doses)
Population
10 43 45 98
(Mi)
Discussion
Question: Keith Klugman inquired about efforts being made to ensure that new vaccines will be efficacious against pneumo-
coccal serotype 1, especially since it is so virulent in South Africa.
Response: Peter Paradiso responded by noting some instances where a vaccine has been efficacious against serotype 1. He
agreed that the positive results are not deemed significant at this point, but continued by emphasizing the fact that studies
have shown that is possible to induce an antibody response against serotype 1. There is a correlate of that response to
functional antibody which could be boosted in the second year of life. Although vaccination against serotype 1 has not been
efficacious at this point, Paradiso provided hope that it could be.
Response: William Hausdorff answered with support for Paradiso’s optimism on the vaccine efficacy against serotype 1. He
went on to consider the immunogenicity and boost response against serotype 1 disease by both the Wyeth and GSK vaccines.
Both vaccines have shown promise against the disease. GSK’s approach to the serotype 1 issue is to consider a shot schedule
with a third shot nine or twelve months of age, or a booster dose.
Comment: Paradiso continued to emphasize the importance of analyzing vaccine efficacy against specific serotypes by
measuring functional antibody response correlates instead of only considering an absolute antibody. Paradiso explained that
some antibodies are more difficult to kill and therefore the assay is different for different serotypes.
Question: Jesus Feris from the Dominican Republic asked Paradiso and Hausdorff about whether there are regulations in
place for administering the 7-, 10- or 13-valent vaccines in place of the primary vaccine during an individual’s vaccination
schedule.
Response: Paradiso added that Wyeth does have data to show that replacing PCV7 mid-schedule with PCV13 does not nega-
tively affect efficacy. He stated, however, that Wyeth has not done testing with PHiD-CV, and was only able to answer in
regards to administration of PCV7 and PCV13.
Question: A response from at least three of the presenters was requested on the importance of the concentration of anti-
bodies and the effects of an adjuvant. Specifically, if an appropriate adjuvant was added to the existing conjugate vaccine,
would it be possible to add an additional seven serotypes to the vaccine to create a much better overall response?
Response: On behalf of PATH, Mark Alderson responded that it would be very difficult to have a new adjuvant approved for
use in a pneumococcal vaccine administered to infants and young children.
Comment: The gentleman who asked the question commented that we know what PCV7 does and we know that we want
higher antibody production. Although we do not want to delay licensure of new vaccines, he felt that the addition of an adju-
vant seems like it should be tried on a parallel track with protein vaccines.
Response: Alderson agreed from a practical and scientific standpoint. However, he noted that the regulatory authorities
have made licensure very difficult to obtain, and that while working with children, especially infants, introduction of a novel
adjuvant to pneumococcal vaccines will be very difficult.
Response: Paradiso agreed with Alderson’s response. He also added that Wyeth did try to enhance conjugate vaccines with
3DMPL, monophosphoryl lipid A. The testing on mice provided increased efficacy, however it did not with children. While
this is just one example, it remains important that the regulatory authorities tend to be very conservative when it comes to
vaccines for infants.
Comment: Hausdorff commented on GSK’s history creating vaccines around adjuvants. Although GSK has succeeded in
doing this for vaccines, including the malaria vaccine and the human papillomavirus vaccine, it remains a very arduous task
to complete adjuvant studies since animal tests are often not predictive. For children, there are other options, like conjugate
or protein vaccines that will work, too. For adults, GSK has considered adjuvants more seriously, but there is still a lot of
risk involved.
Cost-effectiveness of Vaccination
Cost-effectiveness analyses show that the pneumococcal conjugate vaccine (PCV) is a cost-effective — and in certain
contexts, a cost-saving — intervention. In this session, Dr. Anusha Sinha of the New Jersey Medical School (UMDNJ) assessed
three cost-related questions for pneumococcal disease: first, is PCV effective? Second, is PCV cost-saving in certain settings?
And finally, how can health and economic analyses inform country decisions to introduce PCV?
Is PCV cost-effective?
Based on the reviewed analyses of its direct effects, it is estimated that PCV would save from 262,000 to 470,000 lives per
year, with the latter factoring in a very limited herd protection effect. The analysis of the 72 GAVI eligible countries shows
that the cost-effectiveness ratio ranges from $22 per disability-adjusted life year (DALY) averted to $100 per DALY averted.
From this, Sinha noted that the vast majority of studies [of the 7-valent pneumococcal conjugate vaccine (PCV7)’s direct
effects] fall within the World Health Organization’s benchmarks for highly cost-effective interventions.
Sinha et al 2007
Current Situation
Between 2000 and 2007, coverage for the DPT3 and measles vaccines increased from 54 percent to 82 percent. Coverage
hit a plateau between 2006 and 2009. Nshimirimana noted that 80 percent of the children in Africa who are not vaccinated
live in eight countries that are large and where achieving coverage is difficult — including Nigeria, Ethiopia, the Democratic
Republic of the Congo, Tanzania, Kenya, Uganda, Mozambique and Niger — and that 80 percent implementation for DPT3 could
be achieved with a well-targeted initiative in these countries.
A 2007 evaluation in nine of the countries showed promising results. Significantly, in all districts, three out of the five
components of the RED approach were implemented, and the RED strategy was being applied in the nine countries evalu-
ated. However, there were still many issues to be resolved. Health centers in the districts were not managing to reach the
entire community. Eighty percent of the health centers said they had district inspectors come in the last three months, but
inspector feedback was only provided to half of the health centers. Also, regular meetings with the community were held in
two-thirds of the health centers, but they saw that data management and use of data needed to be improved.
Challenges of implementing RED remain. These include weak health delivery and immunization systems, inadequate financial
and human resources for the implementation of the RED approach in many countries, and challenges with maintaining gains
in increasing routine immunization coverage while continuing to introduce new vaccines and ensuring long-term financial
sustainability and coordination mechanisms between various donors and governments.
To address the challenges of RED implementation, the RED guidelines were revised. The new RED Field Guide reflects renewed
interest in reaching all target populations and strengthening linkages with communities. It also included a systemic data
analysis system to monitor routine immunization trends for action and performance indicators at the district level. Country
adaptation rules were added to fit local needs and realities and to support intensive capacity building at the district level.
Lastly, a RED monitoring tool with core and supplemental indicators was also included in the new RED Field Guide.
A proposal to fund the implementation of conjugate meningitis A vaccines has been approved by the GAVI Alliance, which
some Ministers of Health have made a commitment to implement. In addition to the meningitis A vaccine, HPV and malaria
vaccines are also currently in development.
Vaccine Surveillance
From data collected between 2005 and 2008, Hib Impact studies conducted in seven countries — Benin, Burkina Faso, Ghana,
Mali, Rwanda, Senegal and Uganda — showed that Hib is a very good vaccine and is having a positive impact. In addition,
surveillance is being conducted on the pneumococcal vaccine with Netspear.
Rotavirus surveillance began in 2006 and was coordinated by WHO and health ministers. Preliminary (2 year) data indicates
that 42 percent of acute diarrhea hospitalization is due to rotavirus. The results to date are inconclusive, but more informa-
tion should be available regarding this surveillance later this year.
Conclusions
Nshimirimana remains positive about the outlook of WHO/AFRO programs. Plans include using the revised RED guidelines and
RED monitoring tools to address the large number of un-immunized children and reach the 80–90 percent coverage goal;
completing the introduction of Hepatitis B and Hib vaccines while preparing for newer vaccine introduction, particularly
conjugate pneumococcal and meningococcal A vaccines; strengthening new vaccine surveillance to measure vaccine impact;
interrupting wild poliovirus transmission by the end of 2009; consolidating the gains in measles mortality reduction and
preparing for the measles pre-elimination goal.
South Africa
Dr. Anne von Gottberg of the Respiratory and Meningeal Pathogens Research Unit (RMPRU), National Institute for Commu-
nicable Diseases (NICD) of the National Health Laboratory Service (NHLS) presented South Africa’s perspective on how they
reached their decision to introduce new vaccines, and how they are preparing to meet the challenges they bring.
South Africa is conducting national surveillance, communicating with all public health, private sector and military labora-
tories, as well as laboratories with access to mining populations. Additionally, enhanced surveillance was conducted at 23
sites across the country to obtain additional clinical data. The data is biased toward the academic centers and large urban
hospitals where the majority of the enhanced surveillance was conducted.
The surveyors visit the laboratories in person, speaking with technologists and technicians to obtain feedback data and
follow-up on cases that have been missed. The surveyors also visit the hospitals to discuss the results with clinicians and
also bring the clinicians into the laboratories.
For the purposes of this surveillance effort, the only cases that were included were identified from normally sterile site
specimens. An arbitrary cutoff of 21 days was established so that any specimens within 21 days were counted as one case.
Serotyping methodology and guidelines were used for the interpretation of antimicrobial susceptibility testing.
One limitation of the data collected is that invasive disease only measures a small portion of pneumococcal disease burden,
and only patients who have been admitted to a hospital and have had specimens taken are measured. Also, some doctors
don’t routinely take blood cultures on pneumonia cases, while others take them on most cases presenting pneumonia.
The laboratories must process the specimens they receive and report culture positive cases. At each of these steps, there is
a possibility that all of the information may not be successfully transmitted. A quarterly audit that provides a central data
warehouse containing all cases identified over a period of a year is conducted of South Africa’s public health laboratory
system to improve surveillance. It is estimated that 70–80 percent of pneumococcus cases identified in these laboratories
are reported.
This surveillance provides information about age-specific incidence rates, and data from 2006 shows that children less
than one year of age are at greatest risk of pneumococcal disease, with the young to middle-aged adult group that are
HIV-infected being at high-risk as well. About one-third of the cases reported could be qualified or specified as meningitis.
Almost two-thirds of the cases are bacteremia and the vast majority of these are pneumonia and bacteremia and about 10-20
percent are bacteremia without focus.
In children less than five years of age, serotype 14 had the most cases of invasive diseases between 2003 and 2007. Serotype
19A increased between 2003 and 2005 and then stabilized over the next two years. Serotype 1 was higher and then decreased
over the years. Serotype 5 increased in 2004 before decreasing over the next three years.
In children less than five years of age, pooling the data over the time period shows that in South Africa, the most important
serotype is serotype 14 followed by 6B, 6A and 23F. It is worth noting that in children, serotype 1 is not the most important
serotype in South Africa’s population — 19A is still a lead cause of disease in this population.
900
800
700
Number of isolates
600
500
400
300
200
100
0
14 6B 6A 23F 19F 19A 1 9V 18C 4 5 8 15B 3 12F 16 9N 13 34 29 10A 10 15A 7C Other
Serotype
Unlike the incidence rate of all invasive pneumococcal disease, which drops precipitously among children less than one
year to the older children, with serotype 1, older children have similar rates of disease as infants. In older children, serotype
1 disease is associated with HIV-negative children. Serotype 5 presents a more classic view of disease incidence in South
Africa, with a high rate of disease in children less than one year of age and in older adults.
The 7-valent pneumococcal conjugate vaccine (PCV7) serotypes cover about 60 percent of children less than one year old.
The 10-valent pneumococcal conjugate vaccine (PCV10) adds a small amount of coverage and the 13-valent pneumococcal
conjugate vaccine (PCV13) brings the coverage to more than 80 percent. Moving toward older children, PCV10 and PCV13play
a more important role as serotype 1 and other serotypes become more important. Subsequently, moving into the adult age
group these vaccine serotypes becomes less important, as other serotypes cause more disease.
Using the same data but combining the 2–5 year-old age group shows that cross protection of 6A would result in similar
coverage between the two age groups. So, PCV7 plus PCV10 increases coverage because of serotype 1 in the older children, the
2-5 year-old age group, and then there coverage is increased in the younger age group due to the 19A serotype in PCV13.
In HIV-positive children, PCV7, together with serotype 6A, would prevent more disease in the age by percentage than in the
HIV-negative group. Serotype 1 is more important in the HIV-negative group and increases the coverage of PCV10. Again,
serotype 19A plays a more important role in the HIV-positive group.
Using the 2007 and 2008 meningitis Clinical Laboratory Standards Institute (CLSI) break points, the percentage of non-
susceptible isolates causing invasive disease shows that there has been an increase in non-susceptible isolates in South
Africa. Most of this disease occurs in children, with almost 50 percent of children presenting with these non-susceptible
isolates. However, if these isolates are from pneumonia cases, they would be considered susceptible, and high-dose penicillin
would be the preferred antibiotic.
“I think the story in South Africa will be much more complicated than the wonderful
story that the ABC Surveillance has been able to tell in the United States.”
— A n n e Vo n G ot t b er g
In conclusion, serotype 1 disease is important in all children in South Africa, but causes only about 3 percent of disease in
infants. In the older children, it’s mainly associated with HIV-negative children. Seven percent of disease in children less
than two years of age is due to serotype 19A, and of that percentage, more than 50 percent of these isolates will be non-
susceptible to penicillin. In South Africa, invasive pneumococcal disease is associated with about a 75 percent prevalence
of HIV co-infection, so this data really represents HIV co-infection with pneumococcus, which limits the duplication to other
countries. However, with the baseline data, researchers in South Africa should be able to monitor the vaccine introduction.
Mauritius
Dr. Mohammad Issack provided an overview of pneumococcal infection in Mauritius with particular emphasis on invasive
infection, meningitis and bacteremia and antimicrobial resistance.
Mauritius is a small island nation with a fairly stable population of 1.2 million. According to the World Bank, the gross national
income per capita is just over $5,000 USD, making it ineligible for GAVI funding. The Ministry of Health offers free health
services to the entire population, and between 10 and 20 percent of people predominately use private health care facilities.
Bacteriology services from all government health care institutions and from some private clinics are processed at the
Central Health Laboratory. In 2008, the laboratory processed 645 Cerebrospinal fluid (CSF) specimens and nearly 11,000
blood cultures.
Mauritius has a well-established immunization program, although a few changes have been implemented recently so that
all infants receive bacille Calmette-Guerin (BCG) vaccine in the first two weeks of life and then diphtheria, tetanus, pertussis
(DTP), polio, Hib, and hepatitis B vaccines at six, 10, and 14 weeks. Vaccine coverage is very good and in 2007, 87.3 percent of
eligible infants received their third dose of DTP in government health care institutions. Including those who have been vacci-
nated in private practice brings the actual coverage to somewhere between 90-95 percent. The Hib vaccine was introduced
in private practices in 1995, and due to the cost, was not introduced to the routine immunization schedule until March 2006.
There are currently about 16,000 live births each year in Mauritius, so if pneumococcal conjugate vaccine (PCV) were to be
introduced at $30 per dose (which is half the current retail price) and if the three-dose schedule were used, it would cost
about $1.4 million USD. The current vaccine budget is $0.7 million USD.
The meningitis data is based on a prospective recording of all cases since August 1997, the date when agglutination tests on
CSF specimens for Hib, meningococcus and pneumococcus began. The data includes both culture-positive and antigen-posi-
tive cases. The bacteremia and antibody susceptibility testing results are based on a retrospective review of lab results.
n=186 n=58
22%
13% 19%
5% 34%
9%
19%
Hib
1998–2005 Pneumo 2007–2008
Meningo
Strep B
Others
Before and after introduction of Hib, there was a shift in causes in 2007 and 2008 across all age groups. Pneumococcus was
the main cause of bacterial meningitis, accounting for 34 percent of the cases. In the one month to five year age group, Hib
was still the main cause but not by much compared to pneumococcus. Over a two-year period, the absolute number of cases
of Hib and pneumococcal meningitis shows that Hib is decreasing as pneumococcus seems to be increasing.
About one-third of the cases of pneumococcal meningitis occurred in children under one year of age. The vast majority of
these cases occurred in infants who were at least four months old. Hib meningitis seems to occur more commonly between
the months of June and November, during the drier periods of the year when upper respiratory tract infections are usually
more prevalent as well.
Using these figures with their limitations, the annual incidence of pneumococcal meningitis in all age groups would be 0.8
per 100,000 in 2007–2008 and almost 19 per 100,000 in children under 1. Taking into account the blood culture isolates and
the CSF isolates and excluding duplicate isolates from those who had pneumococcus isolated in both blood culture and CSF
in 2007-2008, the rate of invasive pneumococcal disease in the under one population would be 34.7 per 100,000.
In most cases there were no obvious risk factors, but there were a few instances of nephrotic syndrome, past history of
splenectomy, a history of alcoholism or HIV among intravenous drug users.
The rate of penicillin non-susceptibility in blood culture isolates, with non-susceptibility defined by an MIC of 0.12 or greater,
has increased from 8.3 percent to more than 50 percent. Erythromycin resistance and tetracycline resistance also increased
from no resistance out of 24 isolates in 2002 to 20–30 percent resistance in 2007–2008. Among the CSF isolates, resistance
has increased to over 40 percent. Most of the penicillin non-susceptible organisms had MICs between the range of 0.12-1, with
only two isolates with an MIC of 2 or greater.
This data presents limitations, mainly that serotype data of invasive isolates is missing, and there is no data on the burden
of disease of pneumococcal pneumonia. In the case of bacteremia and meningitis, many patients, especially children, tend to
receive antibiotics before admission to hospitals, resulting in false negative CSF and blood culture specimens.
The researchers tried to pick up some cases of partially-treated meningitis by latex agglutination tests, but many studies
have shown that the use of polymerase chain reaction (PCR) or immunochromatographic kits would have picked up even
more cases. The figures don’t include specimens processed in private laboratories; however, this would not have affected
the figure by more than 10 percent.
In conclusion, pneumococcus is the leading cause of bacterial meningitis in Mauritius, partly due to the introduction of the
Hib vaccine in the immunization schedule. However, there has been an absolute increase in the number of cases. Antimicro-
bial resistance has markedly increased in the past 10 years in both invasive and non-invasive pneumococcus isolates. As a
country, Mauritius will continue to consider the cost of the vaccine and the incidence of pneumococcal disease before intro-
ducing PCV. However, availability of local serotype data on invasive isolates would be useful before making this decision.
The DRC is Africa’s third largest country, with a current population estimated at 69 million people with 70 percent of them
living in rural areas. Newborns account for 4 percent of the population. Life expectancy is relatively short — about 45 years
old — and the prevalence of HIV has remained steady at 4.2 percent. DRC is one of the countries with a high infant mortality
rate for children under five years old — an estimated 148 children out of 1000 die before their fifth birthday.
Despite these improvements, estimates issued by the Demography Health Investigation show that from 1993 to 2007, the
mortality rate did not improve much.
Upon analyzing the causes of death in children under five, one of the major causes is pneumonia. “It is estimated that
800,000 children die of pneumonia worldwide — and 132,000 of them die in the DRC of pneumococcal disease,” said Nyembwe.
When examining the distribution of deaths caused by pneumonia from WHO 2002 estimates, DRC is highly impacted.
Studies also reinforce pneumonia’s devastating toll in the DRC. Previous local studies confirmed that Streptococcus pneumo-
niae is the lead cause of bacterial meningitis. The introduction of a vaccine against pneumococcal diseases has the potential
to contribute extensively to a decrease in infant mortality.
Nyembwe emphasized the need to accelerate the introduction of PCV, noting that millions of children will continue dying
from pneumonia without them, and that the introduction of PCV will help achieve the fourth millennium development goal
of reduction of infant mortality rates by two-thirds by 2015. There is already funding to introduce the vaccine through GAVI
support. “So there is no reason to let these children die,” declared Nyembwe.
Today, DRC is officially approved by GAVI and is ready to introduce PCV in 2010. Research on the vaccine’s impact shows that
the vaccine is expected to protect a target number of 2,565,517 children under the age of 12 months in 2011.
The vaccines under consideration contain the highest possible number of serotypes, providing the best protection for most
children. As 2011 approaches, it is highly probable that the DRC will be able to introduce the PCV10, or possibly the PCV13. “It all
depends on our arrangements with GAVI when the vaccine is introduced and what type of vaccine will be available, but we are
looking for the most comprehensive vaccine, to protect children as much as possible,” said Nyembwe.
To prepare for introduction of the vaccine, there are many challenges to overcome. Nyembwe stated the need to strengthen
the bond between community and health services and the need to state the case to the decision makers and pediatric special-
ists so they fully comprehend the scope of the disease. Another challenge will be to educate parents to understand the need
for multiple injections and the benefits of immunization.
7-valent (Wyeth)
10-valent (GSK)
Multi-valent
(emerging suppliers)
Vaccins à nouvelle
technologie
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016-
2020
Nyembwe declared the need to spread the benefits of the immunization far and wide. “The strategy is to reach every health
zone — we will cover all of them,” said Nyembwe. To do so, they will need to boost the capacity of the staff, train the nurses
in injection to minimize the rate of wasting the vaccines, strengthen the cold chain capacity to preserve the vaccines and
strengthen the monitoring systems. The financing is already in place. “We have support for purchasing the vaccine; the funds
only need to be released,” said Nyembwe.
Nyembwe concluded the presentation by emphasizing the DRC’s role in communicating their findings and conclusions with
other countries in their region and acting as advocates. He stressed the importance of making an effort to explain to decision
makers and other authorities the burden of pneumonia and the impact it has on African children’s survival. He optimistically
declared his hope for Africa by stating, “We can say that for the coming generation, we will have an Africa free of life-threat-
ening diseases.”
The Gambia
Dr. Richard Adegbola of the Medical Research Center (UK) in the Gambia provided an overview of the Gambia’s plans to intro-
duce a pneumococcal vaccine in June 2009, and the implications of introducing the current approved form of the vaccine
— PCV7 — which is poised to save lives, though it covers only 63 percent of isolates in the Gambia.
Plans for the Gambia’s introduction of routine PCV7 immunization were at an advanced stage, with the rollout scheduled for
June 2009. The plans also included the future change to PCV10 or PCV13 once they were approved and available for use in the
Gambia. The schedule would fit into the EPI program with three doses at ages two, three and four months.
The primary issue with the Gambia is the fact that the
coverage of serotypes by PCV7 is only at 60 percent. This In spite of the lower serotype coverage, the high incidence
is not to say that the vaccine should not be used. “We of pneumococcal disease in the Gambia means the number
have the data to show that this is a lifesaving vaccine of pneumococcal cases prevented is much higher than
so it should be used,” said Dr. Adegbola. “The question what would be prevented in the U.S. or Australia. This
to be considered is whether or not the vaccine they are makes the implementation of the latest approved version
using is the best vaccine available for their community. of PCV very urgent. “If you wait for another year, many
The current vaccine that is licensed has 60 percent children that could have survived are going to die,” said
coverage, while PCV10 coverage is 80 percent – but is Dr. Adegbola.
not available for use in the Gambia at the moment.”
Dr. Adegbola mentions that there are issues with using Figure 23 | Expected Impact of 7-valent Vaccine in The Gambia
a vaccine of lower valency in a setting like Gambia, and Exceeds that in US and Australia
the issue is not being overlooked. To further validate
the use of PCV7 and determine the vaccine’s full impact, 180
Cases Prevented / 100,000
the committee considered the impact of community- 160
Burkina Faso/Togo
Burkina Faso and Togo lie in the meningitis belt, which is characterized by a pronounced incidence of bacterial meningitis.
Local epidemics are exclusively caused by meningococci, while high incidence during the dry season is due to both meningo-
cocci and pneumococci. Dr. Judith Mueller, of Agence de Médecine Préventive spoke on the implications for Burkina Faso and
Togo being situated on the meningitis belt, and how that affects PCV efficacy and introduction.
Burkina Faso is a country with relatively low HIV prevalence, similar to Togo. Estimates for HIV are less than five percent in
pregnant women. Burkina Faso is expected to introduce a meningococcal conjugate vaccine in the next one to two years. Given
the high incidence of specific non-pneumococcal meningitis incidence and need for the meningococcal conjugate vaccine,
Burkina Faso has not applied or shown interest for pneumococcal vaccine introduction.
An exhaustive surveillance study of pneumococcal meningitis has been going on since 2002 in both Burkina Faso and Togo.
Results for 2007–2008 showed that the annual incidence rate was around seven per 100,000 — less than the 17 per 100,000 that
was reported from Burkina Faso and Ghana earlier in this decade. In the meningitis belt, incidence is particularly high in older
children and adults (two-thirds of cases occur in children younger than five) and serotype 1 is the leading serotype. There is a
very high lifetime risk for meningitis — one in 125 — and a high case fatality ratio at 30–50 percent.
Dr. Mueller assesses that PCV7 could prevent about 30 percent of pneumococcal meningitis cases in children under five and
12 percent in persons over five years of age. These percentages would be 50 percent to 75 percent with serotype1-containing
vaccines. According to her assessment, Mueller finds that PCV7 does not have a notable impact on pneumococcal meningitis
in this population, while recognizing the limitations of the projection. “In the meningitis belt, we really lack serotype-specific
data on bacteraemia, pneumonia and also carriage.” She concludes that this data is needed to compare the socioeconomic
burden and benefits of possible vaccination strategies. She suggests that moving forward, a specific vaccination strategy may
be necessary for the meningitis belt.
Discussion
Question: Dr. Charalambous from London, UK commented that if we introduce PCV, the population will expand, and with any
expansion you require an improved infrastructure. Charalambous asked if Anusha Sinha had taken into account the burden
and the actual costs in her models.
Response: Sinha answered that her analyses accounted for the macro picture changes in demography and capital demands
that such expanding populations require. She also pointed out that traditionally-growing populations with large numbers of
productive adults have been seen by economists as being positive attributes of a society.
Response: Deo Nshimirimana noted that we need to re-evaluate the way we are advocating for vaccine use. He held up Hib
vaccine advocacy efforts as an example, noting that that the Hib Initiative’s efforts were well organized and successful and
resulted in the vaccine being widely adopted, despite early opposition to the vaccine based on the widespread belief that its
introduction would prove too costly.
Question: The gentleman followed up his question by asking if there was a specific advocacy strategy for Africa. He asked
whether, instead of just targeting the Minister of Health and the Minister of Economic Affairs, would be a good idea to target
parliamentarians and other government officials.
Response: Nshimirimana said that that was a good idea and that we should use all opportunities to advocate, including
among parliamentarians.
Question: Dr. Zar from South Africa recommended that in South Africa there should also be targeted immunization of HIV-
infected children, since a huge burden of pneumococcal disease lies in that population. She asked whether there are any
cost-efficacy analyses of targeted immunization strategies as a prelude to a more general immunization strategy.
Response: Thomas Cherian solicited the response of Jose Santos from the National University of Mexico. Santos responded
by saying that the targeted approach has been more empiric, rather than based on real evidence or economic analysis in the
case of both rotavirus and PCV. He noted that when the rotavirus vaccine was first licensed in Mexico (the first country to
license it) and there weren’t enough resources to actually implement the vaccination program, they targeted communities
with a predominantly indigenous native population. He commented that rotavirus is a ‘democratic’ virus that doesn’t affect
only poor people, but children of all economic strata. For the pneumococcal vaccine the same approach was initiated in target
populations and three years since this strategy, there is still a lack of a good analysis of that intervention. He mentioned that
while there is no question that the intervention was beneficial, he is still unsure of the impact on the community at large.
Santos also commented on the economic crisis’ impact on the acceleration of vaccine introduction in middle-income coun-
tries, noting that vaccine costs will impact whether or not they can accelerate the introduction of vaccines in their countries.
They will have to rethink the economics, and while they are all convinced of the value of rapid introduction of PCV, they
will have to consider targeting specific populations as a means to convince policymakers and the population at large of its
worthiness.
Comment: Ron Dagan commented that every country has the resources for vaccine implementation, but it is a matter
of prioritizing the vaccine over other issues and convincing policymakers that it is an easy and important thing to do. He
mentioned that he would like to discuss ways to convince politicians that what they have in hand now with the vaccine is the
highest priority for the next year.
Comment: Adegoke Falade spoke on the situation in Nigeria. Nigeria has made some progress in the last two years on
health care, particularly primary healthcare. Falade commented on the importance of continuing to conduct advocacy and to
demonstrate to policymakers that the introduction of pneumococcal vaccines should be a high priority.
Reponse: Falade thanked him for the ideas and mentioned that they should explore these strategies. Falade invited members
from the National Primary Health Development Agency to comment.
Comment: Amadu Nuhu, the immunization officer for Nigeria, commented that the ground work has already begun on new
vaccine initiative introduction. He mentioned that according to their road map, they hoped to introduce both the pentavalent
and pneumococcal vaccines in the first quarter of 2010.
Comment: Steven Black commented on the cost-saving discussion. He mentioned that he believes that health ministers and
governments need to decide on cost-effective interventions and ‘rate’ interventions relative to other public health measures
or activities they can undertake. He noted that they don’t immunize children to save money, but rather they use public health
interventions that are cost-effective to improve the quality of lives and save lives, and that establishing the criteria that a
vaccine or an intervention needs to be cost-saving falsely raises expectations for most public health interventions.
Response: Sinha responded by pointing out that vaccines are valuable purchases for our societies. Referring back to Till
Bärnighausen’s presentation from the day before, she commented that she’s hopeful for a time in the near future when
vaccines will be considered wise investments.
Much progress has been made in the field of pneumococcal disease prevention. Vaccines have been developed and sufficient
evidence has mounted about the vaccine’s efficacy to convince policymakers to introduce vaccines in their countries. Few
diseases have had such strong evidence in support of the implementation of vaccines to prevent them. Yet there is still the
hurdle of getting policymakers on board to implement these vaccines in the countries that need them the most. This section
will explore both the accomplishments and the challenges to spreading the benefits of pneumococcal disease prevention.
Bravo urged that this is a critical time to raise awareness and advocate for action, noting that we have the evidence, and
PACE is seeking tangible and quantifiable change. Given the current circumstances, Bravo made the case that this change is
possible, noting that with the reason, the evidence and the science to back us up, we cannot fail.
The GAVI Alliance’s mission is to save lives and protect peoples’ health by increasing access to immunization in poor coun-
tries. GAVI is led by four strategic goals: strengthening the capacity of health systems, accelerating the uptake of underused
and new vaccines, increasing the predictability and sustainability of financing and trying to add innovation and value as a
public/private global health partnership.
The work plan is broken down into four areas: ensuring sufficient and appropriate vaccines, securing donors willing to
fund, helping ensure informed country decisions and spurring countries to introduce the vaccine. When all of these areas
overlap, sustainable introduction is achieved. GAVI acts as a platform for these four areas to come together to create
sustainable introduction.
Parliamentarian’s Perspective
The British Parliament established the All Party Parliamentary Group (APPG) on Pneumococcal Disease Prevention in the
Developing World that deals with the issues of pneumococcal disease in the UK and the world. The Honorable Jim Dobbin of
the House of Commons gave the perspective of the Parliament on pneumococcal disease and their current activity with the
newly-formed APPG on Pneumococcal Disease Prevention in the Developing World.
3. Recommendation three was that the Depart- Dobbin noted that the report proved to be a significant step in the
ment for International Development (DFID) continuing fight against pneumococcal disease. The APPG launch
and Advance Market Commitment (AMC) part- brought together key attendees, including 19 ambassadors and
ners publish regular updates on the progress high commissioners, seven British Parliamentarians, two represen-
of the pneumococcal AMC program. tatives from WHO, four representatives from the pharmaceutical
industry, seven experts on pneumococcal disease in the devel-
4. Recommendation four urged AMC donors and oping world and 23 representatives from nonprofit organizations
other international governments to consider and civil society. The event also generated two letters from the UK
supporting further research initiatives to Prime Minister to the group and to the chairman, congratulating all
discover and develop vaccines for adults who of the members for their work.
are also affected by pneumococcal disease.
Another big success of the report was that the follow-up to this
5. The fifth recommendation was to reduce report was written by the Secretary of State for International Devel-
the delay for the vaccine to reach the opment himself — Douglas Alexander. Dobbin noted his hope that
developing world — a delay that is alleged the APPG and other groups would engage with each other, experts
to be about 15 years. and researchers. The APPG is also planning to meet regularly with
parliamentarians who focus on other disease areas to learn from
6. The final recommendation was for govern- each other and see where the common problems are.
ments and agencies to take this opportunity
to educate people on the signs, symptoms Dobbin offered praise to GAVI, PneumoADIP, the Sabin Vaccine Insti-
and risk factors for pneumococcal disease. tute, the World Health Organization, PACE, governments and NGOs,
as well as members of the audience for their research.
He stated that there had been much progress on pneumococcal vaccine introduction in the past five years. Awareness,
supply issues and financing had all significantly improved. Action for accelerating the introduction of PCV in Africa had
already begun. South Africa was leading — with pneumococcal vaccine scheduled to be introduced in April 2009. South Africa
was also the first country in the region to embark on the accelerated measles control initiative that resulted in measles
mortality reduction throughout the continent.
As for the specific goal of PCV, Okwo-Bele emphasized the need to go beyond the goal of having “X” number of countries
introducing the vaccine. He emphasized the need to look at the actual implementation of the pneumococcal vaccine and
corresponding pneumococcal disease reduction as part of the larger strategy to reducing mortality in children. He also noted
that there is an important need to make financing available for those countries that need the vaccines the most, and urged
everyone to support the mission of GAVI.
Dr. Ciro de Quadros of the Sabin Vaccine Institute closed the session by noting that Africa can indeed implement pneumo-
coccal vaccines as demonstrated by their work in eradicating smallpox and significantly reducing measles mortality. He
commented on the tremendous impact of the South African measles initiative, as well as his own public support for it.
He reemphasized that Africa can introduce PCV when they have support, noting the complications presented by the economic
‘meltdown.’ He also stressed that priority that must be given in the next six months to polio eradication, noting that if we fail
with this endeavor, it will be more difficult to implement other health initiatives.
Cynthia Whitney of the U.S. Centers for Disease Control and Prevention (CDC) praised the quality of the presentations and
the information. She remarked that she had never been at a meeting where the presentations and the information made the
need to take care of a problem so clear. She spoke on behalf of CDC — noting that she is very proud to be a co-sponsor of
the meeting.
Fred Were, of the Kenya Paediatric Association, conveyed his excitement at being at the Symposium. He noted the discrep-
ancy of the colors on the map of Africa, which indicated the different stages of vaccine introduction for various vaccines. In
light of this, he stated that he was pleased to see his colleagues working to ‘make those colors uniform’ in front of him. Were
also expressed his delight in having two independent country societies — the South African Pediatric Society and the Kenya
Paediatric Society — co-hosting a meeting of this magnitude.
Orin Levine of PneumoADIP commented that the session should have been renamed “A Call for Continuing Action.” He
remarked, “In fact when the score card is written on the performance of regional offices and countries, the Afro regional
office and the African countries are going to get the A+. You’ve already started taking action.” He praised Africa’s quick
response to GAVI’s decision to support pneumococcal vaccines. “We just have to take our hats off to you. You’re unsung
heroes, and we appreciate that and look forward to more of it,” said Levine. He concluded the session thanking those advo-
cating for pneumococcal vaccines and noting his shared hope that the next meeting will focus on the documentation of the
direct impact of the vaccines.
SPEAKERS
Richard Adegbola Akira Homma Michel Nyembwe
MRC Laboratories, The Gambia Fiocruz, Brazil Ministry of Health, Democratic
Republic of Congo
Olu Akinyanju Mohammad Issack
Sickle Cell Foundation, Nigeria Central Health Laboratory, Mauritius Jean-Marie Okwo-Bele
World Health Organization, Switzerland
Mark Alderson Hope Johnson
PATH, United States PneumoADIP, United States Peter Paradiso
Wyeth, United States
Till Bärnighausen Tatu Kamau
Africa Centre For Health And Population Ministry of Health, Ethiopia Jon Pearman
Studies, South Africa GAVI Alliance, Switzerland
Keith Klugman
Lulu Bravo Emory University, United States Meena Ramakrishnan
University of the Philippines, Philippines PneumoADIP, United States
Orin Levine
Thomas Cherian PneumoADIP, United States Anushua Sinha
World Health Organization, Switzerland New Jersey Medical School, United States
Shabir Madhi
Ron Dagan University of the Witwatersrand, Anne von Gottberg
Soroka University, Israel South Africa National Institute for Communicable
Disease, South Africa
Ciro de Quadros Judith Mueller
Sabin Vaccine Institute, United States Agence de Médecine Préventive, France Fred Were
Kenya Pediatric Association, Kenya
Jim Dobbin David Murdoch
House of Commons, United Kingdom University Of Otago, New Zealand Cynthia Whitney
Centers for Disease Control and
William Hausdorff Deo Nshimirimana Prevention, United States
GlaxoSmithKline, Belgium World Health Organization,
Democratic Republic of Congo