Map | A Partnership for Children's Health Search:  Advanced Search
Home General Information Country Support Board Task Forces Resources Media Center
Immunization Forum GAVI Updates Immunization Focus Board Documents Policy & Technical issues Other websites CVI Forum
Current Issue Archives Subscriptions About Immunization Focus
Printer-friendly format

Quick Reference:

What is GAVI?

GAVI Fact Sheets


Board Documents


Press Releases

Immunization Information

Immunization Forum
Latest Issue

GAVI Update

Immunization Focus - the GAVI quarterly


En Français

Rotavirus vaccines: what next for the countries that need them most?

The withdrawal of a vaccine against a major diarrhoeal disease has raised the stakes for vaccine developers everywhere. Phyllida Brown reports

ROTAVIRUS is one of the biggest killers among diarrhoeal diseases, with an estimated death toll of 600,000 a year - more than one child every minute. A vaccine against the disease was licensed in the United States in 1998. But in October last year the vaccine was withdrawn by its marketing company, Wyeth-Lederle, after it was linked with cases of intussusception, a serious and occasionally fatal blockage of the bowel, in U.S. infants. The consequences - both for those who need rotavirus vaccines and for those who make them - have been profound. Trials of all rotavirus vaccine candidates have been held back and made more complex and costly, forcing back the timeline for reducing the global burden of the disease by years. Here, Immunization Focus finds out what happens next, and what other vaccines are in the pipeline.

Diarrhoeal dehydration: treatment is good, but prevention would be better

Where does the U.S. decision leave developing countries?

The vaccine, known as RotaShield* or RRV-TV, had been tested for efficacy in the United States and Finland before its licensing, but, aside from a successful trial in Venezuela, little was known about whether it could protect children in developing countries where children are typically infected earlier in life and with a wider range of rotavirus strains. So trials had been planned in several countries with a high rotavirus burden, including India, Bangladesh and South Africa. Those trials were put on hold when the data on intussusception emerged in the United States (1,2).

Doctors and researchers from developing and industrialized countries met earlier this year at the World Health Organization to agree how to move forward (3). They called for trials of new candidate rotavirus vaccines as soon as possible.

Importantly, they said that wherever possible, these trials should be done concurrently in low-income and industrialized countries.

  But they also left the door open for trials of RRV-TV itself, under two specific conditions: that trials would monitor babies for signs of intussusception and treat them promptly, and that the manufacturer would guarantee to supply the vaccine for general use if the results were good.

In reality, however, the options are more limited. Wyeth-Lederle is not currently distributing RotaShield anywhere, either for trials or for sale. Wyeth’s Peter Paradiso says that the company has "not given up" on RotaShield’s potential, but neither is it willing to go further without clear indications from regulatory and health authorities in high-burden countries that they would actually want the vaccine following successful trials. "You don’t do trials with a vaccine that nobody would ever use," says Paradiso.

What are the benefits and risks of the vaccine?

The greater the threat of rotavirus in a child’s environment, the greater the potential benefit of the vaccine to that child. In India, for example, about 140,000 infants die of rotavirus disease every year, or one in every 200. In Bangladesh, the figure is 20,000, a comparable death rate to India’s.

Global distribution of annual rotavirus deaths
Source: (4)

Assuming that RRV-TV had given 80 per cent protection from rotavirus-related death in Bangladesh, it could have saved 13,000 lives a year in the nation’s immunization programme alone, even taking account of the fact that not all children would receive the vaccine, estimates David Sack, director of the ICDDR,B Centre for Health and Population Research in Dhaka.

In the United States, rotavirus kills far fewer children - up to 40 a year - although it causes about 50,000 hospitalizations (4). During the period that RRV-TV was available, about 1 million American infants were immunized, and one child died of vaccine-related intussusception.

The final information on the risks of intussusception will not be known until data are complete this summer, but the U.S. Centers for Disease Control and Prevention (CDC) examined the early data and found that the risk was significantly higher in immunized than in un-immunized children within the first two weeks after receiving the first and second doses (2).

In the first week after the first dose, the risk was 25-fold higher in immunized children. In a setting where child deaths are rare, this risk was deemed unacceptable, and the U.S. Advisory Committee on Immunization Practices (ACIP) withdrew its recommendation for use of the vaccine. Perhaps surprisingly, there was no formal discussion in the committee of what an acceptable risk-benefit ratio should be, says Paul Offit, an ACIP member and a rotavirus specialist at the Children’s Hospital of Philadelphia. The committee did, however, point out that its conclusions might not apply in other countries where the risks and benefits might be different (2).

Box 1: Intussusception

What is intussusception?

A blockage of the bowel in which one segment of the intestine folds inside another - like a shirtsleeve being pulled inside out

Does it kill?

Death from intussusception is rare if people have prompt access to treatment. Data from hospitals in African and Asian countries report that between 3% and 26% of cases are fatal, but the data may be incomplete

What causes it?

"Natural" intussusception may be triggered by infections or developmental factors. No one knows how rotavirus vaccines might trigger the condition, but animal strains of the virus seem more likely to cause it than human strains

How common is it?

Few large-scale studies have been done. In the United States natural intussusception affects up to 70 babies in every 100,000 every year; in developing countries, the existing figures suggest lower rates, though some cases may go undetected

Who suffers from it?

Infants, usually aged between 3 and 9 months; more common in well-nourished infants and in boys

Source: (3)

If a vaccine is not safe enough for the United States, is it safe enough for anyone?

Paediatricians and researchers are split into two broad camps. For some, the death toll from rotavirus is simply too great to allow a rare adverse effect to prevent immediate further and careful testing of a vaccine that might prevent hundreds of thousands of deaths a year. Bernard Ivanoff at the World Health Organization, who is responsible for coordinating the agency’s work in vaccine development against diarrhoeal diseases, is personally sympathetic to this view. "Of course, it would be easier for me to say, forget this vaccine," he says. "But it might be five years before we have another one. If we have a vaccine now that can protect people, we could prevent more than 2 million deaths in that time; that’s a concrete reality."

Duncan Steele, director of the South African Medical Research Council’s Diarrhoeal Pathogens Research Unit at the Medical University of South Africa, was involved in the planning of a trial of RRV-TV that was stopped. "Personally, I think that we should have been allowed to continue, albeit with very careful monitoring for intussusception," he says. Claudio Lanata at the Institute for Nutritional Investigation in Lima, Peru, says that clinicians in the city had said that they would still consider using the vaccine if there could first be a large study to assess the risk of vaccine-related intussusception.

For others, it is inconceivable that a vaccine considered too risky for the United States could be tested anywhere else. Although the mathematics of risk and benefit clearly indicate that the vaccine could benefit Bangladesh, says Sack, the political reality is that it would not be acceptable. Imagine the newspaper headlines, he says: "It was not safe enough for Americans, but it’s OK for Bangladesh." Sack gathered a meeting of physicians in Dhaka to discuss the prospects for future trials and the view, he says, was that "there is an urgent need for a successful vaccine, but that it will be difficult to move ahead with RRV-TV".

Paradiso at Wyeth-Lederle agrees. "Regardless of risk and benefit, there are concerns about a vaccine that has known side effects," he says. Parents cannot know their own child’s absolute risks of getting severe rotavirus disease, he says, but they will know that the vaccine carries a risk, however small.

New vaccines in the pipeline

So far, most candidate vaccines against rotavirus have been based on live, weakened animal strains of the virus. These animal strains were used at first, in part, because they grew easily in cell cultures.

Researchers also had evidence that vaccines based on animal strains would protect against human strains. RRV-TV itself is based on a strain from the rhesus macaque. Merck has a candidate based on a bovine strain known as WC3 (see Table 1). To increase the breadth of protection, these strains have been grown in culture together with the most commonly-found rotavirus strains from humans, to make recombined, or reassortant, viruses that also stimulate specific immunity to these human strains. Merck’s candidate has already been tested in Finland and the United States and Timo Vesikari, at the University of Tampere, and others, are involved in plans to test the vaccine in a large trial in these two countries which would monitor both efficacy and safety.

Table 1: Rotavirus candidate vaccines in advanced development

Manufacturer Vaccine type Status

Merck & Co.

Oral vaccine based on five bovine-human reassortant rotavirus strains Safety and efficacy studies in almost 2000 children completed; large safety/efficacy study in Finland and USA is planned to start soon

Glaxo SmithKline (formerly SmithKline Beecham) in partnership with Avant Immunotherapeutics

Known as 89-12 (or "RIX4414")

Oral vaccine based on single, weakened human rotavirus strain

Small safety and efficacy studies completed. Proposals for large trials in both industrialized and developing countries are under discussion

Source: (4)

Vaccines based on human strains may now be attracting more interest. Natural rotavirus infection does not seem to cause intussusception, so some researchers reason that vaccines based on human strains may be less likely to cause it. Glaxo SmithKline (formerly SmithKline Beecham), in partnership with Avant Immunotherapeutics, a company in Boston, has a candidate vaccine, 89-12, based on a human strain. Glaxo SmithKline has said that it will consider doing trials of this vaccine concurrently in both industrialized and developing countries, a significant signal that it recognizes the urgency of the problem and the need for data that all countries, including those with a high disease burden, can learn from.

There are many rotavirus strains that infect humans, but four are known to be common worldwide. RRV-TV and Merck’s candidate both induce protective antibodies against these common strains. Importantly, in India and Bangladesh, there are additional strains that infect a significant number of children - a finding that may be important for vaccine designers (5). Vaccines based on two strains of rotavirus taken from healthy Indian children are being tested for safety in the United States and have attracted the interest of Indian vaccine manufacturers (6).

However, even with vaccine candidates already under development, it is likely to be around five years and possibly as many as ten years before all the results of trials will be known or before a product can be licensed. Add to this the time that is likely to elapse before prices of any registered products fall to globally accessible levels, and the reasons for paediatricians’ frustrations are obvious.

Box 2: Paying for vaccines

Compared with the now remarkably low costs of the traditional vaccines such as diphtheria, pertussis and tetanus, rotavirus vaccines are expensive. RotaShield was priced at $38 per dose, putting it beyond the reach of most low-income countries. Nevertheless, says Ivanoff at WHO, a high initial price must not be a reason for delaying product development that is likely, eventually, to benefit high-burden countries. He cites the falling cost of vaccination against hepatitis B, once about $25 per dose, but now below US$1 a dose within the WHO’s Expanded Programme of Immunization, as grounds for confidence that prices of future rotavirus vaccines would eventually fall. Companies in high-burden countries, such as India, may produce and market future rotavirus vaccines at lower cost than in the industrialized nations.

Meanwhile, research on the second generation of rotavirus vaccines continues. The industry is exploring various options for the longer-term future. For example, researchers are considering developing rotavirus vaccines to be delivered through the nose, says Paradiso. It is hoped that these intranasal vaccines will stimulate an immune response at the body’s mucosal surfaces - where the virus replicates - without the risk that the virus would also trigger intussusception in the mucosa of the gut. Other researchers are trying to develop vaccines based on a part of rotavirus rather than on the whole, live virus. These would be injected rather than swallowed and would be unlikely to carry any risk of intussusception.

The impact on the vaccine industry - down but far from out

No one doubts that rotavirus vaccine developers’ jobs have become more difficult because of intussusception. "The experience with rotavirus vaccine will set a new bar on the size of vaccine trials," says Offit. Researchers estimate that at least 60,000 participants will be needed, and possibly as many as 1 million, for trials to be able to pick up the risk of intussusception. To date, field trials of candidate vaccines against other diseases have rarely involved more than 40,000 people, and usually they have been much smaller.

Waiting game: children like these, in a Vietnamese kindergarten, may be in secondary school before the next rotavirus vaccine is licensed

Beatrice De Vos, director of clinical development in the paediatrics unit of Glaxo SmithKline in Belgium, believes the whole framework for developing rotavirus vaccines has now changed. Now, she says, when the data for any new candidate vaccine is on the regulator’s desk, "somewhere in the head of the person looking at the dossier will be the data from RotaShield". She believes that, rather than simply proceeding along a standard development plan, companies now have to negotiate the development process with the regulatory authorities step by step. The impact may spread beyond rotavirus to other vaccine development programmes, says Alan Shaw, executive director of virus and cell biology at Merck. "This has had a profound impact on how we are going to have to conduct trials in future," he says. "You have now got to do your ’post-marketing surveillance’ before you market a vaccine, and that is expensive." De Vos agrees that conditions have been made tougher. "It looks as though there is a tendency to go for zero risk with vaccines, and that may be impossible," she says. "A baby is at risk just breathing air." For example, a trial might detect no intussusception in 200,000 infants, but the 200,001st child could be the first to suffer the condition: does that mean the vaccine is unsafe? Nevertheless, DeVos remains hopeful. "We are moving ahead."

Given the expected problems with regulators in the industrialized countries after RotaShield, some public health officials think the best hope lies with researchers and manufacturers in high-burden countries, who may be able to develop their own rotavirus vaccines within approval frameworks that are acceptable worldwide. This would be possible, for example, in India, says Julie Milstien in the WHO’s Vaccines and Biologicals division. India already makes and regulates vaccines that are sold to the Expanded Programme on Immunization and its infrastructure for quality control has been fully developed for a number of years.

Paradiso at Wyeth-Lederle says that the cost of developing and testing vaccines has been rising steadily as regulatory requirements have increased. Realistic estimates range between US$200 million and US$400 million for a product, he says, and costs may now spiral further as trials grow larger and their capacity to monitor all participating infants for intussusception is ensured.

Nevertheless, the company has no intention of abandoning its work on rotavirus vaccines. But, says Paradiso, "it seems important for us to move on to new candidates, in addition to evaluating the future of RotaShield. Our overall commitment to vaccine development is still very high, and this has not affected our outlook." That, at least, is good news for tomorrow’s children.

Key references

1. Morbidity and Mortality Weekly Report 48 (27); 577-581

2. Morbidity and Mortality Weekly Report 48 (43); 1007

3. WHO 2000. Future directions for rotavirus vaccine research in developing countries. In press. Contact Dr Bernard Ivanoff for more information:

4. Parashar U. and others. Rotavirus. Emerging Infectious Diseases 4 (4)

5. Ramachandran M. and others. Unusual Diversity of Human Rotavirus G and P Genotypes in India. Journal of Clinical Microbiology . 1996 February; 34 (2) 436-9

6. The Jordan Report 2000. Accelerated Development of Vaccines . National Institute of Allergy and Infectious Diseases. Available via

Immunization Focus May 2000 - Contents

[ star_int ]   Contact us | Guestbook | Copyrights | Illustrated site